Metabolic engineering of Corynebacterium glutamicum for
L-lysine production on silage
Dissertation
zur Erlangung des Grades
des Doktors der Naturwissenschaften
der Naturwissenschaftlich-Technischen Fakultät III
Chemie, Pharmazie, Bio- und Werkstoffwissenschaften
der Universität des Saarlandes
von
Andreas Neuner
Saarbrücken
2012
Tag des Kolloquiums: 18.12.2012
Dekan: Prof. Dr. Volkhard Helms
Berichterstatter: Prof. Dr. Elmar Heinzle
Prof. Dr. Gert-Wieland Kohring
Vorsitz: Prof. Dr. Uli Müller
Akad. Mitarbeiter: Dr. Björn Diehl
Men love to wonder, and that is the seed of science.
Ralph Waldo Emerson
Für meine Eltern, Daisy und Lulu
Danksagung
Mein besonderer Dank gilt Herrn Professor Dr. Elmar Heinzle für die Überlassung des
Themas, das mir entgegengebrachte Vertrauen und die sehr gute wissenschaftliche
Betreuung meiner Arbeit.
Herrn Professor Dr. Gert-Wieland Kohring danke ich für die Bereitschaft zur
Begutachtung dieser Arbeit.
Großer Dank gilt auch der Fachagentur für Nachwachsende Rohstoffe (FNR) für die
finanzielle Unterstützung, sowie Herrn Professor Dr. Roland Ulber für die Koordination
des Projektes.
Sina Balbier, Ines Wagner, Sabrina Lange, Marcel Lambert und Thomas Böse danke ich
für ihren experimentellen Beitrag, den sie während ihrer Forschungs-, Bachelor-, Master-
und Diplomarbeiten geleistet haben. Für die Unterstützung bei der Analytik danke ich
besonders Michel Fritz. Mein Dank geht auch an Veronika Witte, Dr. Klaus Hollemeyer
und Robert Schmidt. Ein herzliches Dankeschön geht an Susanne Kohring für ihr
unermüdliches Engagement in der Arbeitsgruppe.
Allen Mitarbeitern des Instituts für Technische Biochemie möchte ich für ihre große
Hilfsbereitschaft, die schöne Zeit und die einzigartige Atmosphäre danken. Dr. Konstantin
Schneider danke ich sehr für seine Unterstützung und die tolle Zusammenarbeit. Ein
besonders herzlicher Dank gilt Susanne Peifer, für die sehr hilfreichen Diskussionen und
Anregungen sowie für das Teilen der „Doktoranden-Suite“. Vasileios Delis, Tobias Klein,
Sabrina Schmeer und Meike Höfner, sowie Verena Schütz und Yongbo Yuan danke ich für
die schöne gemeinsame Zeit in der aus Arbeitskollegen Freunde geworden sind.
Meinen Freunden Nanu, Alin, Muhammed, Jimmy und Abdul: vielen Dank! Dr. Thomas
Pleli danke ich für seine Unterstützung und Motivation!
Mein besonderer Dank gilt meinen Eltern für ihre unendliche Unterstützung in allen
Lebenslagen und dafür, dass sie immer an mich geglaubt haben. Daisy und Lulu danke ich
für die schöne Zeit außerhalb des Labors.
Table of contents
Table of contents
List of symbols and abbreviations ......................................................................................................... 1
Abstract ................................................................................................................................................... 5
Zusammenfassung ................................................................................................................................... 6
Scope and outline of the thesis ................................................................................................................ 7
Chapter 1 ................................................................................................................................................. 9
Chapter 2 ............................................................................................................................................... 19
Chapter 3 ............................................................................................................................................... 34
Chapter 4 ............................................................................................................................................... 60
Chapter 5 ............................................................................................................................................... 87
Chapter 6 ............................................................................................................................................... 91
References ............................................................................................................................................. 99
Supplementary material....................................................................................................................... 112
Curriculum vitae .................................................................................................................................. 132
Abbreviations
1
List of symbols and abbreviations
Symbols
c Concentration [mol L-1]
µ Specific growth rate [h-1]
qs Specific uptake rate [mmol g-1 h-1]
qp Specific product production rate [mmol g-1 h-1]
T Temperature [°C]
t Time [h] or [min]
U Unit [µmol min-1]
Yx/s Biomass yield [g g-1]
Yp/s Product yield [g g-1]
OUR Oxygen uptake rate [mmol L-1 h-1]
CPR Carbon dioxide production rate [mmol L-1 h-1]
RQ Respiratory quotient
DO Dissolved oxygen [%]
Abbreviations
2-OXO 2-Oxoglutarate
3PG 3-Phosphoglycerate
ABU α-Aminobutyrate
AcCoA Acetyl-Conezyme A
ADP Adenosine diphosphate
Ala Alanine
Abbreviations
2
AMP Adenosine monophosphate
ATCC American type culture collection
ATP Adenosine triphosphate
BHI Brain heart infusion
Bp base pairs
BSA Bovine serum albumin
CDW Cell dry weight
Cit Citrate
CM Complex medium
dld gene encoding D-lactate dehydrogenase
DM Dry Mass
DNA Deoxyribonucleic acid
DTT Dithiothreitol
E4P Erythrose 4-phosphate
F6P Fructose 6-phosphate
F1,6P Fructose 1,6-bisphosphate
FAD+ Flavine adenine dinucleotide oxidized form
FADH Flavine adenine dinucleotide reduced form
fbp gene encoding fructose 1,6-bisphosphatase
G6P Glucose 6-phosphate
G6PDH Glucose 6-phosphate dehydrogenase
GAP Glyceraldehyde 3-phosphate
GAPDH Glyceraldehyde 3-phosphate dehydrogenase
gapX gene encoding glycerinealdehyde 3-phosphate dehydrogenase
Glu Glucose
GDP Guanosine diphosphate
GTP Guanosine triphosphate
Abbreviations
3
HPLC High pressure liquid chromatography
Kan Kanamycin
KanR Kanamycin resistance
Lac Lactate
LB Luria Bertani
ldhA Gene encoding NAD+ dependent lactate dehydrogenase
lldD Gene encoding L-lactate dehydrogenase
LYS Lysine
lysC Gene encoding aspartokinase
lysCfbr Feedback resistant aspartokinase
malE Malic enzyme
MQ Menaquinone
NAD(P) Nicotinamide adenine dinucleotide (phosphate) oxidized form
NAD(P)H Nicotinamide adenine dinucleotide (phosphate) reduced form
NH3 Ammonia
OAA Oxaloacetate
OD Optical density
ORI Origin of replication
PCR Polymerase chain reaction
PEP Phosphoenolpyruvate
PPP Pentose phosphate pathway
Psod Promoter of superoxide dismutase
PTS Phosphotransferase system
Pyc Gene encoding pyruvate carboxylase
PYR Pyruvate
R5P Ribose 5-phosphate
Rpm Rounds per minute
Abbreviations
4
RT Room temperature
S7P Sedoheptulose 7-phosphate
sacB Gene encoding levansucrase of Bacillus subtilis
SOC Super optimized broth
sod Gene encoding superoxide dismutase
SUC Succinate
Taq DNA polymerase from Thermus aquaticus
TCA Tricarboxylic acid
Tet Tetracycline
tkt Gene encoding transketolase operon
XYL Xylose
Abstract
5
Abstract
Currently the essential amino acid L-lysine is mostly produced using traditional substrates
like glucose and molasses. In this work, a computer based approach and molecular biology
techniques were applied in order to investigate the potential of L-lysine overproduction
with C. glutamicum on renewable substrates, in this case silage and silage juice. Based on
elementary mode analysis, several target genes in the feedback resistant C. glutamicum
lysCfbr strain, including D-lactate dehydrogenase (dld), pyruvate carboxylase (pyc), malic
enzyme (malE), fructose 1,6-bisphosphatase (fbp) and glyceraldehyde 3-phosphate
dehydrogenase (gapX), were overexpressed. Substantially re–designing the metabolism
yielded mutants with good growth characteristics, complete substrate consumption,
reduced byproduct formation coupled with increasing L-lysine yields on synthetic and
natural silage juices. This combination of mutations, beneficial for the use of
gluconeogenic substrates and sufficient anabolic reduction power, yielded a robust and
stable strain, C. glutamicum SL. With a total L-lysine carbon yield of around 10% at
growth rates of µ = 0.35 ± 0.01 h-1 on grass and corn silage juices with no further
supplementation and hardly affected by low oxygen supply, this strain proves the
suitability of bio-based L-lysine production.
Zusammenfassung
6
Zusammenfassung
Für die Produktion der essentiellen Aminosäure L-Lysin wurden bisher traditionelle
Kohlenstoffquellen wie Glukose und Molasse benutzt. Unter Zuhilfenahme
bioinformatischer und molekularbiologischer Methoden wurde die Einsatzmöglichkeit von
Silage und Silagepresssaft als Fermentationssubstrat für eine Lysinproduktion mit C.
glutamicum geprüft. Basierend auf der durchgeführten Elementarmodenanalyse wurden im
Ausgangsstamm C. glutamicum lysCfbr die D-Laktat Dehydrogenase (dld),
Pyruvatcarboxylase (pyc), Malatenzym (malE), Fruktose-1,6–bisphosphatase (fbp) und die
Glycerinaldehyd-3-phosphat-Dehydrogenase (gapX) überexprimiert. Die Umstrukturierung
des Zentralstoffwechsels lieferte Mutanten mit guten Wachstumsraten, einem breiteren
Substratspektrum, geringerer Nebenproduktbildung und erhöhten Lysinausbeuten auf
synthetischen und natürlichen Silagepresssäften. Durch die Kombination an Mutationen,
die eine bessere Verwertung von glukoneogenen Substraten und eine ausreichende
NADPH Versorgung ermöglicht, entstand der robuste und stabile Stamm C. glutamicum
SL. Mit Lysinausbeuten von 10% bei einer Wachstumsrate µ = 0.35 ± 0.01 h-1 auf
verschiedenen Silagepresssäften auch bei reduzierter Sauerstoffversorgung, zeigt dieser
Stamm ein großes Potential für Lysinproduktion auf Silagen als nachwachsendem
Rohstoff.
Scope and outline of the thesis
7
Scope and outline of the thesis
The focus of this thesis is on metabolic engineering of L-lysine production with C.
glutamicum using silage and silage juice as substrate. Contrary to commonly used
substrates, silage juice contains high amounts of lactate and other organic acids, requiring
smart engineering in order to satisfy the demand of carbon building blocks and redox
power needed for L-lysine production. While C. glutamicum wild type is not able to grow
on silage juice, elementary mode analysis provided useful targets for genetic engineering
of the metabolism. Once applied, growth and L-lysine production steadily improved. The
utilization of the various carbon sources was investigated in detail, linking their
consumption to strain specific parameters like growth rate and L-lysine production. This
work has the potential to pave the way for a larger spectrum of biotechnological products,
evaluating silage as a renewable, sustainable substrate for biotechnological production.
In Chapter 1 we provide a general introduction on the necessity of replacing commonly
used petrochemical feedstock with bio-based feedstock for the production of chemicals, in
this case the amino acid L-lysine.
The best candidate for this application, C. glutamicum, is introduced in Chapter 2 as a
production platform for various chemicals, especially amino acids. Computational methods
and bioinformatics combined with targeted genetic modifications provide a very effective
toolbox for target identification and metabolic engineering of this organism.
The design and performance of created mutants on synthetic silage juice is described in
detail in Chapter 3, showing the improved substrate spectrum of the created mutants, the
improved growth behavior and L-lysine production.
In Chapter 4 we focus on different silages and methods of processing as well as
drawbacks related to formation of inhibitory compounds during the treatment of the
feedstock. The performance of engineered mutants on different silage juices and under
different cultivation conditions is described in detail.
Pretreatment methods for corn silage juice and potential applications are described in
Chapter 5.
Scope and outline of the thesis
8
In Chapter 6 we summarize the findings presented in Chapters 3-5 and discuss their
potential use and contribution to a sustainable, bio-based economy along with an outlook
on future research in this field.
Chapter 1
9
Chapter 1
General introduction
Chapter 1
10
General introduction
With over 7 billion people and the numbers still growing, the efficient use of our natural
resources is one of the top challenges of our time (Pinstrup-Andersen and Pandya-Lorch,
1998). Economic growth is greatly depending on safe, sustainable resources for industrial
production of chemicals and transportation fuels. Currently, only a minor fraction of the
chemical industries output is based on renewable raw materials (Lichtenthaler and Peters,
2004). For the major part, the chemical industry relies on fossil resources, coupled with
environmentally damaging production processes, waste and very bad eco-efficiency
(Mecking, 2004). The constant search for oil and the subsequent exploitation of these fossil
resources have negative effects on environment and society. The constantly increasing
global demand for petroleum based goods will soon exceed the exploitation, resulting in a
peaking of oil prices and the depletion of these resources in near future (Bentley, 2002).
Pollution, immense costs for waste disposal and a rapid rise in the costs of mineral oil are
increasing the pressure to make chemical production more eco-friendly. For the
environment, non-degradable or non-recyclable products as well as toxic byproducts
represent a serious ecological challenge (Wackernagel and Yount, 2000). Furthermore, the
emission of carbon dioxide and other greenhouse gases in the atmosphere and the related
global warming calls for attention (Yu et al., 2008). All these facts are pointing in the same
direction, a shift from fossil to bio-based raw materials. An attractive alternative feedstock
for the production of chemicals and fuels is biomass. Green biomass, including agricultural
and forestry residues, energy crops, grasses and woody plants being one of the largest
sources of renewable, sustainable resources worldwide (Clark, 2007; Hall and Scrase,
1998; Hoffert et al., 1998), can be used to produce biogas and liquid fuels to generate
electricity or to create bio-based products like plastics, resins and fine chemicals. The
production of plant biomass is vast, sufficient to match the demand of food and feeds with
the demand for the production of chemicals and fuels (Dale, 2003; Ericsson and Nilsson,
2006). Optimal use of agricultural and forestry residues, waste biomass and the cultivation
of dedicated crops for non food production is a vantage point for the bio-based industry
(Hill, 2000). Breeding to increase the yield performance of plants and even alter the
cellular composition of these raw materials is already practiced (Boehmel et al., 2008).
One of the major benefits of using bio-based chemicals instead of petroleum based ones is
the closed carbon cycle. While storage of CO2 in fossil raw materials has taken millions of
years, the CO2 generated by the combustion of bio-based fuels can be incorporated in
Chapter 1
11
biomass via photosynthesis in a much shorter period of time (van Maris et al., 2006). Bio-
based chemicals can be biodegraded into CO2 and water which are both used during
photosynthesis for the regeneration of biomass (Figure 1.1).
102 years
RawMaterials (Oil / Gas)
Biotechnology
Figure 1.1: Schematic overview of the carbon cycle for A: the biotechnological route,
or B: the petrochemical route. Adapted from van Maris et al. (2006).
But a change in the feedstock can only occur, if the technological basis for its industrial
processing is altered. Linking the production of chemicals and biotechnological processes
in general with the rapidly emerging bio energy industries is a promising starting point
(Kamm and Kamm, 2004; Kamm and Kamm, 2007). Analogous to the petroleum refinery,
the concept of the biorefinery has emerged (Figure 1.2). The processing facilities of a
biorefinery are centered on agricultural or farming units, where biomass is converted into
various compounds that are fed into various product lines.
Chapter 1
12
Figure 1.2: Schematic concept of a biorefinery. Based on
http://www.nrel.gov/biomass/biorefinery.html.
The natural complexity and divergency of biomass can be useful for the production of
multiple products, maximizing the value derived from biomass. Compared to conventional
production, in a long run, sustainable production systems should be more profitable, since
they use materials and energy more efficiently reducing also the associated waste
production. This can be achieved, by producing bio-based products that are durable, less
toxic, recyclable and biodegradable, while performing well, compared to their
conventional counterparts. An industry is truly sustainable when it is economically viable,
environmentally compatible and socially responsible.
Role of biotechnology in sustainability
One of the core tools facilitating the shift from fossil to bio-based production is
biotechnology. Also known as white biotechnology, the industrial biotechnology is often
able to comply several green chemistry principles. Multi step chemical synthesis can be
replaced with a single step synthesis of desired products with a lot less energy and material
input as well as reduced waste generation, organic solvent free media, selective catalysis
and biodegradable products (Sijbesma and Schepens, 2003). Furthermore, white
biotechnology enables the synthesis of products that cannot be synthesized chemically
Bio
degr
adab
ility
Lignin content
Wat
er c
onte
nt
Chapter 1
13
(Gavrilescu and Chisti, 2005). With a growing number of organisms entering the post
genomic era and advances made in the fields of enzyme and metabolic engineering as well
as proteomics and bioinformatics, the choice for suitable microorganisms and enzymes for
bioconversions is facilitated. Screening for novel organisms, isolated from extreme
environments (Madigan and Marrs, 1997) or the use of metagenomics (Lorenz et al., 2002)
helps discovering new biocatalysts. In vitro evolution of such enzymes can further improve
the enzyme activity, achieving much higher conversion rates, than in nature. The
combination of high titer, tailor made production strains and engineered enzymes can for
example improve biocatalysis for biomass hydrolysis, one of the key components for
biorefineries. Biorefineries represent a complex system of ecological technologies for the
comprehensive material and energetic utilization of such renewable raw materials,
produced in a sustainable way, where useful products are synthesized without producing
any new waste (Kamm and Kamm, 2004). Especially in Europe, most green biorefineries
use silage instead of fresh, green biomass, guaranteeing a constant substrate supply
throughout the year. This ensures a more decentralized system, with the farmers being
more integrated by running the silos, providing social sustainability for the whole concept
(Hanegraaf et al., 1998).
Traditional biotechnological processes have been used for years. Microbial production of
enzymes, antibiotics or amino acids are only a few examples (Hermann and Patel, 2007).
Regarding amino acids, the history of microbial amino acid production started in the late
1950’s, when Dr. Kinoshita discovered that C. glutamicum is a superior amino acid
producer. Until then, amino acids were either extracted or produced by chemical synthesis.
Today, over 2 million tons of amino acids per year are produced by microbial
fermentations using genetically engineered strains, with a market growth of around 10%
per year (Leuchtenberger et al., 2005). In the era of bioeconomy, producer strains are no
longer engineered only towards a desired product but also regarding the ability to
efficiently use alternative, renewable substrates. In this context, we decided to investigate
the use of silage and silage juices for the production of L-lysine.
Chapter 1
14
Silage
A more effective way to preserve the biomass and make it available throughout the year is
ensiling (McDonald et al., 1991). Ensiling or silaging is the process of preserving plant
material in undried and anaerobic conditions, either in a storage silo or wrapped in plastic.
It is an old method of moist forage preservation, accounting for more than 200 million tons
of biomass stored annually in Europe and the US (Weinberg and Muck, 1996). After the
harvest, the green plant material is chopped and left to wither up to a dry mass content of
approximatively 30% and finally pressed to remove remaining oxygen and assure
anaerobic conditions. Based on natural fermentation, lactic acid bacteria convert a large
fraction of the present water-soluble carbohydrates into lactic acid. As a result, the pH
decreases, inhibiting detrimental anaerobes. If the forage is too moist, not allowing a
sufficient pH decline, clostridia, which ferment lactic acid to butyric acid and amino acids
to ammonia, might become active. This process is referred to as “secondary or clostridial
fermentation” (Weinberg and Muck, 1996). When anaerobic conditions are not guaranteed,
the presence of oxygen enables various aerobic spoilage microorganisms to become active
(Woolford, 1990). The most frequently involved microorganisms in aerobic deterioration
of silages are yeasts and acetic acid bacteria, followed by bacilli, moulds and enterococci
(McDonald et al., 1991; Spoelstra et al., 1988). After providing anaerobic conditions, lactic
acid bacteria develop and become the predominant microbial population. The produced
lactic acid is an interesting product for the manufacture of biodegradable polymers
(Södergard, 2002). The downside of the lactic acid production is the recovery cost (Datta et
al., 1995). Instead of extracting lactic acid from the silage juice, we try to use it as an
additional carbon source for L-lysine production (Neuner and Heinzle, 2011).
Chapter 1
15
Table 1.1: Different types of silage and their composition.
Plant Grass Corn Sugar Beet
Dry mass (DM) 35 – 45% 25 – 30% 30 – 35%
Protein 135 g kg-1 80 g kg-1 110 g kg-1
Lactic acid ~200 g kg-1 ~130 g kg-1 ~160 g kg-1
WSC* + Starch ~45 g kg-1 ~300 g kg-1 ~30 g kg-1
Fiber 240 g kg-1 200 g kg-1 210 g kg-1
Potential 5 – 15 DM/ha/a 15 – 20 DM/ha/a 10 – 15 DM/ha/a
Yield 172 m3/t MM 202 m3/t MM 67 m3/t MM
Biogas 54% methane 52% methane 72% methane
* water soluble carbohydrates
On a large scale, silage is either used for the production of biogas (Table 1.1) or as fodder
at farming units (Kromus et al., 2004). With meat production stagnating and milk
production reaching higher productivities per cow, the percentage of utilized grassland will
decrease and grassland will become a surplus problem for agriculture. In the next decade,
50-100 million hectares of land in the EU may become available for purposes other than
food crop production (Weiland, 2003). This opens up new substrate opportunities for
biotechnical engineering with two conceivable options. The first one would be the
integration of L-lysine fermentation into the process of a green biorefinery (Figure 1.3) and
the subsequent use of the purified product, L-lysine as a fine chemical.
Chapter 1
16
Figure 1.3: The product pipeline of a biorefinery. Adapted from Kromus et al.(2004)
This would be an additional, waste free step in the production pipeline of a biorefinery,
contributing to a sustained value enhancement, with all the residues being used for biogas
production. Offering considerable environmental benefits and an additional income source
for the farmers, biogas production from biomass is of growing importance. The biogas
yield from cow and pig manure is low due to two factors, insufficient organic dry matter
Chapter 1
17
content (2-10%) and most of the energy rich substances have been digested by the animals
(Amon et al., 2007). This leads to the conclusion, that the use of manure as the sole
substrate for biogas production is not economical and makes the use of co-substrates
necessary. An evaluation of modern biogas plants showed that corn and grass silage are the
most applied co-substrates in agricultural biogas plants. Over 80% of the plants are
operated with simultaneous fermentations of manure and silage. Figure 1.4 displays the
application frequency of various co-substrates for biogas production (Weiland, 2003).
Figure 1.4: Application frequency of various co substrates (Weiland, 2003).
Wastes from food industries, greases and fat containing residues as well as residues from
restaurants, markets and the municipal area are mainly used in large, centralized plants,
because of the needed pretreatment and the required technological equipment. Silage and
byproducts from agricultural industries as well as energy crops are therefore very
interesting alternatives as co-substrates and for mono fermentations in smaller farming
units, since they do not require sophisticated special treatment before use. Besides that, the
fertilizer quality of the digestate is high (Angelidaki et al., 2009).
Corn si
lage
Grass s
ilage
Cereal
residu
es
Grease
Lawn c
utW
hey
Food r
esidu
es
Vegeta
ble W
astes
Potato
Potato
pulp
0
10
20
30
40
50
60
70
80
App
licat
ion
frequ
ency
[%]
Chapter 1
18
The second option would be the supplementation of fodder with L-lysine without any
product purification. Proteins are essential components in feeds. Many amino acids can be
synthesized by the animal itself. In the case of the essential amino acids, the animal relies
upon the amino acids supplied by the food. Vegetable feeds have an own characteristic
amino acid spectrum, which differs from that of the animal. L-lysine supplementation is
dramatically increasing the biological value of most plant proteins, since they tend to be
poor regarding L-lysine (Belitz, 2001; Mitchell and Smuts, 1932). This issue is discussed
since the 1930’s (Mitchell and Smuts, 1932) and it is still unsolved today (Krajcovicova-
Kudlackova et al., 2005), mainly because of the high prices for the needed
supplementation. In order to increase the feed efficiency, L-lysine rich crops, like soybean
meal are added to traditional animal fodder like corn, wheat and barley. An advantage is
the increased food quality that is associated with the drawback that other amino acids from
soybean meal are not used by the animals and therefore increase the amount of secreted
nitrogen significantly. In terms of fodder, one of the most important aspects is the PER
(protein efficiency ratio). The PER is based on the weight gain of a test subject divided by
its intake of a particular protein (Belitz, 2001). It is an indicator of the protein quality.
Maize, which is the best producer of both calories and protein per acre (Johnson and Lay,
1974), exhibits a PER of 0.85. A L-lysine supplementation of 0.4% increases the PER to
1.08. Further addition of 0.07% tryptophan results in a PER of 2.5. In case of wheat
protein, a 0.2% L-lysine supplementation is raising the PER from the initial value of 0.65
to 1.56, making L-lysine the most used amino acid in fodder supplementation (Belitz,
2001), improving the animals ability to utilize nitrogen strongly enhancing growth
(Leclercq, 1998) and decreases the release of nitrogen into the environment. A carefully
adjusted amino acid supplementation allows a reduction in the nitrogen release of about
60%, attributing L-lysine a considerable ecological relevance to the farming industry
(Kircher and Pfefferle, 2001). Since Corynebacterium glutamicum belongs to the GRAS
category of microorganisms, the L-lysine enriched fermentation broth could be added to
the fodder without any further purification.
Chapter 2
19
Chapter 2
Corynebacterium glutamicum as a platform
for biotechnological production
Chapter 2
20
Corynebacterium glutamicum as a platform for
biotechnological production
Corynebacterium glutamicum is a Gram positive, non-motile, aerobic soil bacterium with a
GC content of 54.1% (Pfefferle et al., 2003). Like Arthrobacter and Brevibacterium, it
belongs to the group of coryneforme bacteria, named after the typically rod shape with
clubbed ends (Figure 2.1). Contrary to the related pathogenic species, Corynebacterum
diphteriae and Mycobacterium tuberculosis, it is non pathogenic, and exhibits the GRAS
status (Burkovski, 2008).
Figure 2.1: Microscopical image of Corynebacterium glutamicum (source:
whymashen.wordpress.com).
It was originally isolated from a soil sample of the Ueno Zoo in Japan (Udaka, 1960). In
the late 1950’s, when Dr. Kinoshita started a research program aimed at obtaining a
microorganism able to secrete amino acids (Kinoshita et al., 2004), the industrial
biotechnology was not aware of the role Corynebacterium glutamicum was going to play
in the future. Starting as a natural glutamate producer, C. glutamicum emerged to one of
Chapter 2
21
the top players in biotechnological production. With a yearly production of over 2 million
tons L-glutamate and 1.5 million tons of L-lysine, amino acids are still on top of all
biotechnological products of C. glutamicum (Kohl and Tauch, 2009). After entering the
post genomic era (Tauch et al., 2002a), the variety of products synthesized but also of
substrates used by C. glutamicum increased dramatically (Becker and Wittmann, 2011).
Being naturally able to metabolize various carbon sources, including different sugars and
organic acids, genetically engineered C. glutamicum strains with extended substrate
spectrum were created, e.g. utilizing xylose (Kawaguchi et al., 2006), starch (Seibold et al.,
2006; Tateno et al., 2007), arabinose (Schneider et al., 2011), glycerol (Rittmann et al.,
2008) and many others. The flexible and robust metabolism combined with advanced
omics technologies (Wittmann, 2010), targeted genetic engineering (Jäger et al., 1992) and
bioinformatics (Neuner and Heinzle, 2011; Schilling et al., 2000; Schuster et al., 1999)
successfully broadened C. glutamicum’s product portfolio. Besides various amino acids
including non proteinogenic amino acids, diamines like cadaverine (Kind et al., 2011;
Mimitsuka et al., 2007) and putrescine (Schneider et al., 2012; Schneider and Wendisch,
2010), vitamins, dicarboxylic acids (Okino et al., 2008), polymers (PHB) (Liu et al., 2007),
higher alcohols (Smith et al., 2010) and fuels (Inui et al., 2004a) are all products of this
versatile key player in industrial biotechnology.
Central carbon metabolism
The central metabolic network of C. glutamicum comprises the pathways of glycolysis,
pentose phosphate pathway, tricarboxylic acid cycle, the glyoxylate shunt and various
anaplerotic and cataplerotic reactions, while the Entner Doudoroff pathway is missing
(Eikmanns, 2005; Kalinowski et al., 2003). Regarding L-lysine biosynthesis, the central
metabolic pathways provide the precursors oxaloacetate and pyruvate, energy in form of
ATP and GTP as well as the redox equivalent NADPH. NADPH is needed as a cofactor
for the biosynthesis of different amino acids, including L-lysine, L-methionine and L-
tryptophan, exhibiting the highest NADPH demand. Especially concerning L-lysine
production, the supply of this cofactor is of special interest, since L-lysine biosynthesis is
coupled to a high demand of 4 mol NADPH per mol of L-lysine. A key characteristic of
Chapter 2
22
superior L-lysine producer strains is a sufficient NADPH supply. In C. glutamicum, two
enzymes of the oxidative branch of the pentose phosphate pathway, glucose 6-phosphate
dehydrogenase and 6-phosphogluconate dehydrogenase as well as the TCA located
isocitrate dehydrogenase and the malic enzyme are NADPH generating reactions (Gourdon
et al., 2000; Marx et al., 1996; Wittmann and de Graaf, 2005). NADPH consuming
reactions are besides L-lysine biosynthesis the growth associated anabolic reactions with a
associated stoichiometric demand of 16.4 mmol NADPH (g biomass)-1 (Wittmann and de
Graaf, 2005).
Glycolysis
The glycolysis is important for the supply of precursors for cellular anabolism. Besides
that, it provides energy in form of ATP. Hexoses like glucose and fructose are metabolized
in the glycolysis and further oxidized in the TCA. In glycolysis, a set of ten reactions is
converting 1 mol glucose to 2 mol pyruvate, gaining 2 mol ATP and 2 mol NADH. Two of
these ten reactions are irreversible and have to be bypassed when gluconeogenic substrates
like lactate and acetate are used (Cocaign-Bousquet and Lindley, 1995; Netzer et al.,
2004). These two bypassing reactions are catalyzed by PEP carboxykinase and fructose
1,6-bisphosphatase (Stryer, 2007). Glycolytic and gluconeogenic activity is strictly
regulated, depending on the organisms needs and available carbon sources. One of the key
regulators in glycolysis/gluconeogenesis is the fructose 1,6-bisphosphatase (Rittmann et
al., 2003). Increased amounts of AMP, indicating low energy levels in the cell have
inhibitory effects on fructose 1,6-bisphosphatase, downregulating gluconeogenesis and
promoting glycolysis (Shiio et al., 1990).
TCA cycle
In addition to supplying essential redox equivalents and energy, the TCA cycle also
provides precursors required for biomass formation. Since these intermediates are
constantly withdrawn for anabolism, an extensive set of different enzymes is responsible
for the anaplerotic replenishment of these metabolites (Petersen et al., 2000). Especially
when overproduction of amino acids belonging to the aspartate or glutamate family is
Chapter 2
23
intended and OAA is constantly needed for their biosynthesis, the anaplerotic enzymes and
the C3/C4 metabolism play a major role (Sauer and Eikmanns, 2005). The enzymes PEP-
and pyruvate carboxylase replenish the OAA pool, in order to maintain the functionality of
the TCA cycle. The C4 decarboxylating enzymes PEP-carboxykinase, OAA decarboxylase
and the malic enzyme replenish the phosphoenolpyruvate and pyruvate pools, all together
forming the so called PEP-pyruvate-oxaloacetate node of C. glutamicum (Eikmanns,
2005). This anaplerotic node represents the link between glycolysis and the TCA cycle. At
the same time, when gluconeogenic carbon sources like lactate, acetate or a TCA cycle
intermediate are used, the anaplerotic node is the starting point of gluconeogenesis.
Pentose phosphate pathway
The general role of the pentose phosphate pathway (PPP) is the supply of anabolic
reducing power in form of NADPH and precursor metabolites like ribose 5-phosphate and
erythrose 4-phosphate, used for the synthesis of aromatic amino acids and nucleotide- and
nucleoside production. It has two branches, the oxidative, irreversible route, where glucose
6-phosphate is converted to ribulose 5-phosphate by three enzymes, with the formation of
2 moles of NADPH per mol glucose (Moritz et al., 2000). The reversible, non oxidative
route, including the enzymes transketolase and transaldolase is not generating any redox
equivalents. The role of the non oxidative route is the isomerization, epimerization and
interconversion of C3, C4, C5, C6 and C7 compounds. The PPP is an alternative to
glycolysis, with a rather anabolic than catabolic role, branching at glucose 6-phosphate and
refueling glycolysis at the levels of fructose 6-phosphate and glyceraldehyde 3-phosphate.
Especially in L-lysine producers, with a higher requirement of NADPH, the flux
partitioning through the PPP is of major interest (Moritz et al., 2000; Wittmann and
Becker, 2007).
Chapter 2
24
Lactate metabolism
While the uptake system for L-lactate has not been unequivocally identified, C.
glutamicum can grow aerobically on L-lactate as a carbon and energy source (Cocaign-
Bousquet and Lindley, 1995; Cocaign et al., 1993; Gourdon et al., 2000; Seletzky et al.,
2006). In a first step L-lactate is oxidized to pyruvate via L-lactate dehydrogenase encoded
by the lldD gene (Bott and Niebisch, 2003; Schluesener et al., 2005; Stansen et al., 2005).
The utilization of L-lactate is mainly regulated by the expression levels of lldD, which was
shown to be organized in an operon with a putative L-lactate permease (NCgl2816).
Compared to cells grown on minimal medium containing glucose, fructose or pyruvate, the
mRNA levels of this operon were 18-fold higher when L-lactate was used, indicating a
carbon source dependent transcriptional regulation (Burkovski, 2008; Stansen et al., 2005).
The genome sequence revealed another gene, dld. The protein encoded by the dld gene
shows 46% sequence identity to a D-lactate dehydrogenase from E. coli and several other
bacteria (Burkovski, 2008). While the lldD is inducible, dld is constitutively expressed, at
very low levels (Scheer et al., 1988). Both dehydrogenases are peripheral membrane
proteins, using menaquinone as electron acceptor. C. glutamicum possesses a third,
soluble, NAD+ dependent lactate dehydrogenase, encoded by the ldhA gene (Stansen et al.,
2005). While the NAD+ dependent lactate dehydrogenase LdhA is used for reoxidation of
NADH under conditions, where the respiratory chain is limiting, the menaquinone
dependent dehydrogenases Dld and LldD are mainly involved in lactate oxidation, when
used as a carbon and energy source (Burkovski, 2008; Eggeling and Bott, 2005). The
formed pyruvate is channeled into the TCA cycle by the pyruvate dehydrogenase complex
and the pyruvate carboxylase. Lactate utilization has also a major impact on expression
levels of the gluconeogenic enzymes PEP carboxykinase and fructose 1,6-bisphosphatase.
Both enzymes activity and mRNA concentrations are 5 fold higher in lactate grown cells
compared to glucose grown cells. This indicates a transcriptional control of these steps in
gluconeogenesis (Burkovski, 2008; Georgi et al., 2008).
Chapter 2
25
L-lysine production and cofactor economy in C. glutamicum
L-lysine production in C. glutamicum is closely linked to various supporting pathways. For
the synthesis of one L-lysine molecule, one molecule of oxaloacetate and pyruvate, 4
molecules of NADPH, two molecules of ATP and one molecule of NH3 are used.
L-lysine is part of the aspartate family of amino acids, consisting of L-lysine, L-threonine,
L-methionine, L-isoleucine and the non proteinogenic amino acid D,L-diaminopimelate,
the direct precursor of L-lysine and an important building block for cell wall synthesis
(Wehrmann et al., 1998). The steps towards L-lysine biosynthesis (Figure 2.2) include the
phosphorylation of aspartate by aspartate kinase and its subsequent reduction to aspartate
semialdehyde. After the condensation step catalyzed by dihydrodipicolinate synthase,
which converts aspartate semialdehyde to dihydrodipicolinate and its reduction to
piperideine dicarboxylate, the pathway may split. Contrary to E. coli, where L-lysine is
synthesized in four different steps, converting piperideine dicarboxylate to D,L-
diaminopimelate, C. glutamicum is able to catalyse this conversion in one step by direct
ammonium incorporatrion by the diaminopimelate dehydrogenase (Bartlett and White,
1985). In depth analysis revealed that C. glutamicum exhibits both variants of D,L-
diaminopimelate synthesis, called the succinylase way and the dehydrogenase way,
respectively (Schrumpf et al., 1991; Wehrmann et al., 1998). Further studies revealed that
flux distributions through the two available routes is mainly dictated by the availability of
ammonium (Misono and Soda, 1980). This allows C. glutamicum and other bacteria,
including Bacillus macerans, which posses this split pathway, to flexibly adapt to
alterations in the nitrogen supply (Wehrmann et al., 1998) (Sonntag et al., 1993).
Chapter 2
26
Figure 2.2: Pathway and regulation of L-Lysine biosynthesis in C. glutamicum. The
enzymes encoded by the genes are ppc – phosphoenolpyruvate carboxylase, pyc –
pyruvate carboxylase, aspC – aspartate aminotransferase, lysC – aspartokinase, asd –
Chapter 2
27
aspartate semialdehyde dehydrogenase, hom – homoserine dehydrogenase, dapA –
dihydrodipicolinate synthetase, dapB – dihydrodipicolinate reductase, dapD –
tetrahydrodipicolinate succinylase, dapC – succinyl-diaminopimelate
aminotransferase, dapE – succinyl-L-diaminopimelate desuccinylase, dapF –
diaminopimelate epimerase, lysA – diaminopimelate decarboxylase, ddh –
diaminopimelate dehydrogenase, lysE – L-lysine exporter, lysG – L-lysine export
regulator. The broken arrow indicates feedback inhibition.
The synthesis of amino acids is strictly regulated, to cover the organisms demand and
avoid unnecessary production. In case of L-lysine, the regulation mechanism is a feedback
inhibition of the aspartate kinase by the endproducts of the pathway, L-lysine and L-
threonine. By binding at the regulatory β subunit, the affinity of the catalytic α subunit for
the substrate aspartate decreases and the biosynthesis is diminished (Stryer, 2007). In order
to overproduce L-lysine, the feedback inhibition must be deregulated. Product feedback
inhibition of allosteric enzymes is an essential issue for the development of highly efficient
microbial strains for bioproduction. Exchange of a single base in the lysC gene, encoding
the aspartate kinase may lead to feedback resistant strains, e.g. the substitution of threonine
with isoleucine at the position 311 in the amino acid sequence of the enyzme (Kim et al.,
2006). The result is a deregulated aspartate kinase and a reduction of the feedback
inhibition and therefore overproduction of L-lysine. In the last 50 years considerable
efforts to deregulate this enzyme from allosteric inhibition by L-lysine and L-threonine
have been made, identifying partially or completely desensitized mutants (Chen et al.,
2011).
Chapter 2
28
Systems metabolic engineering of C. glutamicum
Many industrial strains for the production of useful metabolites have been developed by
classical whole cell random mutagenesis. Repeated random mutation and selection yielded
producer strains with remarkable characteristic qualities (Demain, 2000; Rowlands, 1984).
Although this classical method has enabled great progress in the biotechnological industry,
it has serious disadvantages. Regarding industrially important properties like growth and
stress tolerance, classically derived production strains are often inferior to the wild type
strains and the disadvantage of supplementing the defined medium with essential
metabolites due to the auxotrophy increased production costs (Kiss and Stephanopoulos,
1992). Besides the desired mutations, a lot of uncharacterized mutations might occur, some
being detrimental to the strains performance. Undefined mutations leave the production
mechanism undetermined, an impediment for further rational metabolic engineering. It was
obvious that future success in increasing the performance and yield of already highly
productive strains will depend on the availability of detailed information on the metabolic
pathways and their regulations (Sahm et al., 1995). During the 1980’s, several C.
glutamicum genes from the biosynthetic pathways leading to the aspartate-family amino
acids L-lysine, L-threonine, and L-isoleucine, have been cloned and analyzed (Sahm et al.,
2000). These studies led to a better understanding of the metabolic pathways, but a
complete insight into the complex interactions was impossible, due to the lack of detailed
genetic information. However, many successful examples of strain engineering using
molecular techniques have been reported (Cremer et al., 1991; Ikeda and Katsumata, 1992;
Shiio et al., 1984). A more rational design, based on biochemical and physiological data
obtained from continuous cultures was developed. Implementing these data, a
mathematical formulation based on balancing extracellular substrate consumption as well
as biomass and product formation rates enabled a novel way of analyzing the complex
metabolic network of C. glutamicum (Stephanopoulos, 1999). Developed and applied by
Vallino and Stephanopolous, metabolite balancing was successfully used to determine
metabolic fluxes and identify potential bottlenecks (Vallino and Stephanopoulos, 1993).
However, this approach has certain limitations due to the required constraints. By entering
the post genomic era, the missing information for the rational, engineered development of
industrial C. glutamicum strains finally became available (Ikeda and Nakagawa, 2003;
Chapter 2
29
Kalinowski et al., 2003; Ohnishi et al., 2002). Together with significant progress in the
development of analytical techniques (Wittmann and Heinzle, 1999; Wittmann and
Heinzle, 2001a) the application of 13C isotope labeling (Heinzle et al., 2008; Wiechert et
al., 2001; Yang et al., 2006) and related isotopomer modeling, the prerequisites for more
elaborated metabolic flux analysis, namely an exact cellular composition, existing
reactions and network topology were available. A quantitative knowledge about the in vivo
activity of certain pathways as well as their contribution to the overall activity of the cell
has proven crucial for an in depth understanding of C. glutamicum and for further
optimization of producer strains (Krömer et al., 2004; Sauer and Eikmanns, 2005;
Wendisch et al., 2006; Wittmann and Becker, 2007; Wittmann and de Graaf, 2005;
Wittmann and Heinzle, 2001b; Wittmann and Heinzle, 2002; Yang et al., 2005).
Innovative methods for genetic manipulation (Kirchner and Tauch, 2003; Vertes et al.,
2005) combined with powerful bioinformatical tools (Neuner and Heinzle, 2011;
Radhakrishnan et al., 2010; Schuster et al., 1999; Schuster et al., 2007; Stelling et al.,
2002) for target identification, the overexpression or deletion of genes using recombinant
DNA techniques (Kirchner and Tauch, 2003) is today’s top method for the design and
construction of strains with a desired genotype (Becker et al., 2011; Kind et al., 2010;
Krömer et al., 2006; Wittmann and Becker, 2007).
Targeted genetic engineering
Though various studies are still carried out using plasmids for the expression of genes,the
targeted genetic engineering of C. glutamicum is nowadays primarily performed via
homologous recombination (Kirchner and Tauch, 2003). The genetic construct, containing
a 500 bp homologous region is synthesized via PCR. The genetic modifications include
deletions (Marx et al., 2003; Schwarzer and Puhler, 1991), overexpression (Becker et al.,
2005; Neuner and Heinzle, 2011), point mutations (Kim et al., 2006), start codon exchange
(Becker et al., 2011) and heterologous gene expression (Liebl et al., 1992). All these
genetic modifications and the corresponding genetic constructs are performed via fusion –
PCR. The principle of this so called SOE – PCR is described in Figure 2.3.
Chapter 2
30
Figure 2.3: Principle of the fusion PCR, connecting two DNA fragments.
The stable integration of the modified DNA into the genome is facilitated by the use of
integrative plasmids. These plasmids possess an ORi (origin of replication) for E. coli but
not for C. glutamicum. The plasmids containing the engineered insert are amplified in E.
coli and receive the specific methylation pattern of C. glutamicum. The information for the
methylation is encoded on the pTC plasmid, contained in the used E. coli strain. This
methylation step is dramatically increasing the transformation efficiency, since C.
glutamicum is degrading xenogenic DNA (Bonamy et al., 2003; Tzvetkov et al., 2003).
The purified, methylated plasmids are further used for transformation of C. glutamicum.
The commonly used transformation technique is electroporation (Kirchner and Tauch,
2003; Liebl et al., 1989). Since the integrative plasmids have no ORI for C. glutamicum,
only cells with an integrated plasmid can replicate and form colonies on selective agar
plates. The selection of the 1st recombination occurs via kanamycin resistance. Testing of
the clones is performed via PCR, where the positive clones display both alleles, the wild
type allele and the engineered one. Positive clones can be directly used for the 2nd
recombination. During the 2nd recombination, the wild type allele should be replaced by
the engineered one via homologous recombination. Selection of the 2nd recombination
Chapter 2
31
clones is facilitated by the second selection marker sacB, encoding for the levan sucrose,
an exoenzyme from Bacillus subtilis. This enzyme degrades sucrose to levan, a toxic
substance for C. glutamicum (Jäger et al., 1992). Clones still possessing the pClick int sacB
plasmid can express the levane sucrose and are not able to grow on agar plates containing
sucrose. Only positive mutants that performed the allelic exchange and wild type cells with
the rejected plasmid can grow on sucrose containing plates. The verification of the 2nd
recombination is also done by PCR. The engineered DNA section is sequenced, in order to
exclude undesired mutations. Validation of the mutant strains is carried out by enzyme
tests and cultivation experiments. A schematic overview of the recombination steps for
genome based modifications in C. glutamicum is depicted in Figure 2.4.
Figure 2.4: Recombination steps for genome based modifications in C. glutamicum.
Chapter 2
32
Elementary mode analysis
The engineering of microorganisms for overproduction of desired chemicals involves
simultaneous optimization of multiple objectives like productivity, extended substrate
range and improved tolerance. The achievement of all these objectives is almost
impossible without mathematical modeling and simulations. Different algorithms have
been developed, based on a stoichiometric network comprising all educts, intermediate
metabolites and products, while kinetic parameters are not explicitly required for the
calculations (Papin et al., 2004; Schilling et al., 2000; Schuster et al., 1999; Schuster et al.,
2007). Since more advanced sequencing technologies provided gene sequences of many
different organisms (Benson et al., 2008; Kanehisa et al., 2008; Karp et al., 2007),
metabolic pathway analysis became an irreplaceable component, a basis for rational strain
design. This type of analysis has been successfully applied to various organisms in order to
investigate their metabolic network structure, robustness or regulation, providing a
rigorous basis to systematically characterize certain genotypes and the corresponding
phenotypes (Carlson and Srienc, 2004a; Carlson and Srienc, 2004b; Poolman et al., 2004;
Price et al., 2002; Price et al., 2003; Wiback et al., 2004; Wlaschin et al., 2006). The
elementary mode analysis (EMA) is decomposing metabolic networks of highly
interconnected reactions into uniquely organized pathways. Each pathway is consisting of
a minimal set of irreversible reactions which operate as a functional unit at steady state,
leading from substrate to product. Removal of any reaction in an elementary mode will
automatically disrupt the entire pathway. By eliminating inefficient pathways, engineered
strains can be forced to function only according to efficient pathways, coupled to cell
growth (Pfeiffer et al., 1999; Schuster et al., 1999). Metabolic pathway analysis also allows
the determination of the overall capacity, i. e. the maximal theoretical yield, showing
potential effects of genetic modifications. This information is very useful, since the
economic efficiency and feasibility of the process can be estimated. This also allows the
comparison between different organisms (Krömer et al., 2006). The use of silage and silage
juices as sustainable, renewable carbon sources instead of the commonly used glucose as a
fermentation substrate requires tailored strains, with a broader substrate range. Therefore,
instead of the commonly used, product oriented approach, we applied a similar but more
advanced, product and substrate oriented approach. We adapted our computational method
to the complexity of the substrate, silage. Our focus was on three main carbon sources,
Chapter 2
33
glucose, fructose and lactate. We accounted for possible variations regarding their
concentration, depending on the used green biomass, ensiling techniques and seasonal
factors like temperature, humidity and harvest time. Detailed information and the results of
the applied elementary mode analysis are given in Chapter 3.
Chapter 3
34
Chapter 3
Mixed glucose and lactate uptake by Corynebacterium glutamicum through
metabolic engineering*
*Published as: Andreas Neuner and Elmar Heinzle (2011): Mixed glucose and lactate uptake by Corynebacterium glutamicum through metabolic engineering, Biotechnology Journal, Volume 6, Issue 3, pages 318-329, DOI:10.1002/biot.201000307
Chapter 3
35
Abstract
C. glutamicum ATCC 13032 lysCfbr strain was engineered to grow fast on racemic
mixtures of lactate and to secrete L-lysine during growth on lactate as well as on mixtures
of lactate and glucose. The wild type C. glutamicum grows only well on L-lactate.
Overexpression of D-lactate dehydrogenase, dld, carried out by exchanging the native
promoter of the dld gene with the stronger promoter of the sod gene encoding superoxide
dismutase in C. glutamicum resulted in a duplication of biomass yield and faster growth
without any secretion of L-lysine. Elementary mode analysis was applied to identify
potential targets for L-lysine production from lactate as well as from mixtures of lactate
and glucose. Two targets for overexpression were pyruvate carboxylase and malic enzyme.
The overexpression of these genes using again the sod promoter resulted in growth
associated production of L-lysine with lactate as sole carbon source with a carbon yield of
9% and of 15% during growth on a mixture of lactate and glucose. Both substrates were
taken up simultaneously with a slight preference for lactate. As surmised from the
elementary mode analysis, deletion of glucose-6-phosphate isomerase resulted in a
decreased production of L-lysine on the mixed substrate. Elementary mode analysis
together with suitable objective functions has been found a very useful guide for the design
of strains producing L-lysine on mixed substrates.
Chapter 3
36
Introduction
Raw material supply is a key factor in the production of bulk biochemicals like amino
acids. Traditionally waste materials like molasses, various hydrolysates but also starch, e.g.
from corn, have been used. In countries with moderate climate with their seasonal changes
sustained raw material supply for fermentation is restricted to stabilized materials like
starch or molasses. An alternative method of stabilization of agricultural raw materials is
silage that has been used for a long time for feed preservation. Silage is extensively and
increasingly used in agricultural and cattle industry, since it has a 70% higher nutritional
value than hay (Van Soest, 1982). More recently silage is also used for the production of
biogas (Kromus et al., 2004) and serves such as a renewable resource for the supply of
methane. Two major compounds of silage juice produced from grass are lactate and
glucose (Krotscheck et al., 2004). Lactate is considered primarily responsible for
conservation of the material, i.e. restriction of the growth of other bacteria. It was now
interesting whether such type of raw material could potentially be used to produce more
valuable compounds than methane. Potentially useful products are essential amino acids
like L-lysine or L-methionine which are extensively used as animal feed additives (Ikeda,
2003). C. glutamicum, a gram positive soil bacterium grows aerobically on various
carbohydrates and organic acids as carbon sources (Liebl et al., 1991) and is the most
frequently used organism for industrial production of amino acids (Wittmann, 2010).
Glucose and lactate are co-metabolized (Cocaign et al., 1993). Considerable knowledge
about the genes and enzymes involved in glucose and lactate metabolism is available
(Gourdon et al., 2000; Stansen et al., 2005). Nevertheless, growth on glucose-lactate
mixtures has only been studied marginally and, to our knowledge, no attempt to produce
amino acids from lactate and mixtures of lactate and glucose using engineered strains has
been reported. C. glutamicum has two genes encoding quinone dependent lactate
dehydrogenases, L–lactate dehydrogenase (lldD) and D–lactate dehydrogenase (dld). Both
are peripheral membrane proteins (Schluesener et al., 2005), serving to oxidize lactate
when it is used as a carbon and energy source. LldD is transcribed together with
NCgl2816, coding for a putative permease, in one operon. Additional lactate carriers seem
to exist in C. glutamicum, since the deletion of NCgl2816-lldD operon led to inability to
grow on L-lactate as carbon and energy source, and complementation with a plasmid only
carrying lldD in this deletion mutant recovered growth (Stansen et al., 2005).
Chapter 3
37
Transcriptome analysis of L-lactate grown C. glutamicum showed high mRNA levels of
NCgl2713 and NCgl2965, both encoding putative permeases (Stansen et al., 2005). L-
lactate dehydrogenase is inducible, D-lactate dehydrogenase is constitutively expressed,
but with a very low expression rate (Scheer et al., 1988). Therefore, the constitutive D-
lactate dehydrogenase (dld) of the L-lysine producing C. glutamicum strain lysCfbr is a first
obvious target for overexpression to increase D-lactate utilization of C. glutamicum
(Scheer et al., 1988). The ldhA gene, encoding the NAD+ dependent, soluble lactate
dehydrogenase is used for reoxidation of NADH under conditions where oxidation by the
respiratory chain is limiting and lactate is formed as a fermentation product (Inui et al.,
2004b). Producer strains developed by classical methods often have uncharacterized
mutations, leading to diminished stress tolerance, decreased substrate uptake or
consumption whereas targeted engineering leads to robust production strains (Ohnishi et
al., 2002). Elementary flux mode analysis and extreme pathway analysis (Papin et al.,
2004), (Schilling et al., 2000), (Schuster et al., 1999) can be used to identify potential
targets for directed metabolic engineering. First it permits the estimation of maximum
possible yields (Krömer et al., 2006). Elementary flux mode analysis has also been used to
correlate product fluxes with individual network fluxes to identify genes for
overexpression in C. glutamicum and A. niger (Melzer et al., 2009). Flux balance analysis
was used to study the impact of deletions and varying oxygen supply on C. glutamicum
metabolic performance (Shinfuku et al., 2009). Several differing genome scale network
models of C. glutamicum were recently published almost in parallel (Kjeldsen and Nielsen,
2009; Melzer et al., 2009; Shinfuku et al., 2009).
Chapter 3
38
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
YLys/S [C-mol C-mol-1]
YB
M/S
[C-m
ol C
-mol
-1]
Figure 3.1: Carbon yields of biomass and L-lysine determined by elementary mode
analysis using metabolic network of C. glutamicum (Supporting information, Figure S
3.1 and Table S 3.1). The network includes glucose uptake with PTS as well as via
passive diffusional transport and glucokinase according to Shinfuku et al. (2009). It
also comprises a reversible malic enzyme (open circles – elementary modes with
growth only on lactate; full triangles – elementary modes with growth only on
glucose; open diamonds – elementary modes with simultaneous consumption of
lactate and glucose).
Since an effective NADPH supply has been identified as important prerequisite for
superior L-lysine producers (Wittmann and Heinzle, 2002), the manner in which NADPH
can be furnished during growth on substrates employing gluconeogenic pathways is of
major interest (Cocaign-Bousquet and Lindley, 1995). To check to which extent L-lysine
Chapter 3
39
yield during growth on mixed substrates is linked to an increased flux through the
oxidative branch of the pentose phosphate pathway (PPP) (Wittmann and Heinzle, 2002),
the deletion of glucose-6-phosphate isomerase gene is an interesting target (Marx et al.,
2003).
In this work, elementary flux mode analysis was carried out for L-lysine production
starting from published genome scale networks (Kjeldsen and Nielsen, 2009; Melzer et al.,
2009; Shinfuku et al., 2009). It was used to investigate lactate, glucose and mixtures
thereof as carbon sources for maximal yields of L-lysine. Flux modes representing high L-
lysine production with a minimal enzyme requirement promise to deliver information
which metabolic reactions and pathways are potential targets. This analysis predicted
stimulation of L-lysine production by a combined overexpression of pyruvate carboxylase
and malic enzyme. The present work describes overexpression of D-lactate dehydrogenase,
pyruvate carboxylase and malic enzyme and also the deletion of glucose-6-phosphate
isomerase and its impact on growth and L-lysine production on racemic lactate as well as
on glucose-lactate mixtures.
Chapter 3
40
Materials and Methods
Microorganisms: Bacterial strains and plasmids used in this study, their relevant
characteristic and their sources of reference are listed in Table 3.1. All mutants were
designed and constructed on the basis of the wild type C. glutamicum ATCC 13032
(American Type and Culture Collection) with deregulated L-lysine biosynthesis (allelic
replacement of the lysC gene with a lysCT311I gene) (Kim et al., 2006). The vector used
for introducing the modified genes, plasmid pClik is carrying a kanamycine resistance and
the sacB gene as selective markers. The pClik plasmid has no origin of replication (ori) for
C. glutamicum. Transformation of the organism with the plasmid and selection for the
kanamycin resistance yielded transformants with genome integrated plasmid DNA.
Integration of the plasmid DNA occurred via a single crossover homologous
recombination. The second recombination was detected and selected via the sacB positive
selection system (Jäger et al., 1992). Several sucrose resistant, kanamycin sensitive clones
were tested for the presence of the mutation by PCR. Additionally, sequencing of the
resulting PCR product was done. The primer sequences used for the verification of the
promoter exchange upstream of the dld, pyc and malE genes as well as the corresponding
fragment size for the wild type and the mutant alleles are listed in Table 3.2.
Chapter 3
41
Table 3.1: Molecular biological tools and strains constructed starting from C.
glutamicum ATCC 13032.
C. glutamicum lysCfbr Exchange T311I in the lysC gene Kim et al. 2006
C. glutamicum lysCfbrdldPsod lysC T311I + Exchange of the natural promoter of the
dld gene by the promoter of the sod gene
This work
C.glutamicum
lysCfbrdldPsodpycPsod
lysC T311I + Exchange of the natural promoter of the
dld and pyc genes by the promoter of the sod gene
This work
C. glutamicum
lysCfbrdldPsodpycPsodmalEPsod
lysC T311I + Exchange of the natural promoter of the
dld, pyc and malE genes by the promoter of the sod
gene
This work
C.glutamicum
lysCfbrdldPsodpycPsodmalEPsod∆
pgi
lysC T311I + Exchange of the natural promoter of the
dld, pyc and malE genes by the promoter of the sod
gene + deletion of the pgi gene
This work
E.coli DH5α F- endA1, hsdR17 (vk- mk+) supE44, thi-I λ- recAI
gyrA96 rel A1, ∆ (lac ZYA-argF)U169 F80d lacZ
∆M15
(Hanahan, 1983)
E.coli NM522 supE thi-1 ∆(lac-proAB)∆(mcrB-hsdSM) 5(rK- mK+)
[F’ proAB laclq Z∆M15]
Stratagene
Plasmids
pClik int sacB
vector for integrative, allelic replacement by homologuous recombination, nonreplicative in C. glutamicum, KanR, sacB
BASF
Media: The first preculture was grown in complex medium containing 10 g L-1 glucose, 5
g L-1 yeast extract, 5 g L-1 beef extract, 5 g L-1 polypeptone, 20 g L-1 casaminoacids, 2.5 g
L-1 NaCl and 2 g L -1 urea. For agar plates, 20 g L-1 agar were added. Second preculture
and the main culture were grown in minimal medium (pH 7.2) containing (per litre): 10 g
glucose, 10 g lactate, 16 g K2HPO4, 4 g KH2PO4, 5 g (NH4)2SO4, 300 mg 3,4-
dihydroxybenzoic acid, 10 mg CaCl2, 250 mg MgSO4*7H2O, 10 mg FeSO4*7H2O, 10 mg
MnSO4*H2O, 2 mg ZnSO4*7H2O, 200 µg CuSO4*5H2O, 20 µg NiCl2*6H2O, 20 µg
Na2MoO4*2H2O, 100 µg cyanocobalamin, 300 µg thiamine, 4 µg pyridoxal phosphate and
100 µg biotin.
Chapter 3
42
Cultivation: Single colonies from an agar plate were used as inoculums for the first
preculture, which was grown for 8 h at 30°C in a 100 ml baffled shake flask with 10 ml
medium on a rotary shaker (Multitron II, Fa. Infors AG, 4103 Boltringen, Schweiz) at 230
rpm. Cells were harvested by centrifugation (3 min, 6500 x g, 4°C, Labofuge 400R,
Heraeus, Hanau, Deutschland), washed twice with sterile 0.9% NaCl and used as inoculum
for the second preculture, which was cultivated overnight under the same conditions. Main
cultures were performed in duplicate using 250 ml baffled shake flasks with 25 ml
medium. Cell concentration was determined by a photometer (Novaspec®II, Pharmacia
Biotech, Little Chalfont, UK) at 660 nm. The correlation between cell dry weight (CDW)
and the optical density at 660 nm was determined as described previously (Krömer et al.,
2004).
Cell disruption: Cells were harvested by centrifugation (5 min, 6800 x g, 4°C). A small
amount of glass beads (Ø 0.25 mm) and 1 ml MilliQ were added to the pellet. Disruption
was performed using a bead mill (Retsch, MM301) at maximum speed (30 Hz) for 35
seconds. Cell debris was removed by centrifugation.
DNA preparation and transformation: Oligonucleotide synthesis was done by Sigma
Aldrich. DNA sequencing by GATC, Konstanz, Germany. Chromosomal DNA from C.
glutamicum was obtained using the Instant Bacteria DNA Kit (ANALYTIC JENA, Jena,
Germany). Plasmids from E. coli were isolated using the GFXTM PCR DNA and Gel Band
Purification Kit (GE Healthcare, Chalfont St. Giles, Great Britain). E. coli was transformed
by heat schock (Inoue et al., 1990). C. glutamicum was transformed by electroporation
(van der Rest, 1999),(Tauch et al., 2002b). Cultivation of the C. glutamicum for
electroporation was done in a 10 ml preculture, grown in BHI++ medium, inoculated from
a fresh agar plate. The main culture (BHI++) was inoculated with the overnight preculture
to an optical density of 0.2 at 660 nm. At an optical density around 0.6 400 mg isonicotinic
acid hydrazide, 2.5 g glycine and 0.1 ml Tween 80 were dissolved in 20 ml H2O and added
to the cultivation medium filter sterilized. Cells were harvested at an optical density of 0.8
at 660 nm. Electroporation was performed with parameters set at 25 µF, 600 Ω and 2.5 kV.
Immediately after electroporation, 1 ml BHIS medium was added to the cells and the
suspension was transferred into a 2 ml Eppendorf tube for the following heatshock at 46°C
Chapter 3
43
for 5 minutes. After the heatshock, the cells were incubated at 30°C for 90 minutes, to
allow recovery and subsequently spread on BHISkan agar plates.
DNA manipulations: All PCR reactions were done in a TGradient-Cycler (Whatman -
Biometra®), with FidelyTaq (Fermentas, Hilden, Germany) or Jumpstart RedTaq Mix
(SIGMA) using 10 mM dNTP Mix (Fermentas, Hilden, Germany) and the 1kb DNA
O’GeneRulerTM (Fermentas, Hilden, Germany). Ligations were done with Fast-LinkTM
DNA Ligation Kit (Epicentre Biotechnologies, Madison, Wisconsin, USA). Restriction
enzymes and their buffers were obtained from Fermentas (Hilden, Germany) and used as
recommended by the manufacturers.
Overexpression of the dld, pyc and malE genes: Overexpression of the genes dld, pyc
and malE in C. glutamicum was achieved by cloning the open reading frame (ORF) of the
mentioned genes under the control of the promotor of the sod gene, encoding superoxide
dismutase (NCgl2826). Wild type gene length and length after introduction of the sod
promoter or deletion: dld 903/1095 bp, pyc 926/1118 bp, malE 917/1109 bp, pgi
2223/1820 bp.
Table 3.2: Primer sequences used for verification of the mutant strains and the PCR
fragment size of the wild type and mutant alleles.
Genetic
modification
Primer sequence PCR fragment size
Psoddld Fw: 5’- GATCCTCGAGTCTGATTGCTGCGTCGATC-3’
Rw: 5’-GATCACGCGTCGAGTTGTTCGCGATG-3’
WT: 903 bp
Psoddld: 1095 bp
Psodpyc Fw: 5’-ATCGCTCGAGCTAATTTTTCTGAGTCTTAG-3’
Rw: 5’-
CGATACGCGTGCCTTCACAAAGATGGGGTAAGTC-3’
WT: 926 bp
Psodpyc: . 1118 bp
PsodmalE Fw: 5’-ATCGCTCGAGCTTACCAAGTGGG-3’
Rw: 5’-CGATACGCGTCGTGCATAACTGG-3’
WT: 917 bp
PsodmalE: .1109 bp
∆pgi Fw: 5’-GATCACGCGTATCCCTTCTCCGGCATC-3’
Rw: 5’-GATCTCTAGATCCAGCGACACGAATAATC-3’
WT: 2223 bp
∆pgi: 1820 bp
Chapter 3
44
D-lactate dehydrogenase and malic enzyme assay: For determination of enzyme
activity, exponentially growing cells were harvested by centrifugation (5 min, 7000 rpm,
4°C) and washed three times with working buffer (50 mM ice cold Tris HCl, pH 7). After
resuspension in 1 ml 50 mM Tris HCl, pH 7, cells were disrupted with a bead mill (Retsch,
MM301) at maximum speed for 35 seconds. Cell debris was removed by centrifugation.
Enzyme activity was measured in cell free supernatant. For the D-lactate dehydrogenase,
reaction mixtures of 1.5 ml contained 1.4 ml 0.2 M Tris HCl, pH 7.3, 50 µl
NAD+(100mM) .The reaction was started by adding 50 µl 30 mM D-lactate. The reaction
temperature was 25°C. For malic enzyme, the oxidative decarboxylation of malate
contained 0.1 M KH2PO4 buffer, pH 7.8, 0.6 mM NADP+, 5 mM MgCl2 and 40 mM
sodium L-malate. Enzyme activity was determined spectrophotometrically by measuring
the absorbance change of NAD(P)H at 340 nm. Specific activity was calculated relative to
the protein content of the cell extract. The protein content was quantified using the
Bradford method (Bradford, 1976).The reagent solution was obtained from Biorad.
Negative controls were performed without substrate or cell extract, respectively.
Analysis of substrates and products: Quantification of sugars and organic acids in 1:20
diluted supernatant was done by high pressure liquid chromatography (Kontron
Instruments, Neufahrn, Germany) using a Aminex HPX-87H column (Bio-Rad, Hercules,
Calif.) at 60°C with 7 mM H2SO4 mobile phase and a flow rate of 0.8 ml min-1. Detection
was via determination of refraction index (sugars) or UV absorption at 210 nm (organic
acids). Quantification of amino acids was done as described by Krömer et al. (2005).
Metabolic reaction network: The metabolic network of C. glutamicum was set up for
utilization of glucose and lactate based on the work done by Krömer et al.(2006) , using
data from KEGG database (http:/ www.genome.jp/kegg/metabolism.html) as well as
biochemical and physiological literature (Eggeling and Bott, 2005). It also incorporates
collected knowledge from three recent publications of genome scale metabolic networks of
C. glutamicum scale networks (Kjeldsen and Nielsen, 2009; Melzer et al., 2009; Shinfuku
et al., 2009). It comprises glucose uptake via phosphotransferase system (PTS) and a
putative permease combined with glucokinase (Shinfuku et al., 2009), lactate uptake via a
permease system and lactate dehydrogenase, glycolysis pathway (EMP), pentose
Chapter 3
45
phosphate pathway (PPP), tricarboxylic acid cycle (TCA), anaplerotic pathways including
reversible malic enzyme (Kjeldsen and Nielsen, 2009), glyoxylate cycle, biomass
production, L-lysine biosynthesis pathways and respiratory chain (supplementary Table S
3.1). For ATP production in the respiratory chain a P/O ratio of 2 (for NADH) and 1 (for
FADH) was assumed (Klapa et al., 2003). The precursor demand for biomass formation
was taken as described in the literature (Marx et al., 1996). Water, protons, phosphate and
sulphur were not included in the balances. Putative glucose transport not using the PTS is
described differently in the literature. While Kjeldsen and Nielsen (Kjeldsen and Nielsen,
2009) incorporated only energy driven transport of glucose either at the expense of ATP or
protons, Shinfuku et al. (2009) used a simple permease that was also incorporated in the
model here.
Computational methods: Elementary modes analysis was carried out using “efmtool”, an
open source program available at http:/www.csb.ethz.ch/tools/efmtool. The mathematical
details of the algorithm are described in the literature (Terzer and Stelling, 2008). For each
flux mode, the carbon yields of biomass (Yx/s) and L-lysine (YLys/s) were calculated as
percentage of the substrate carbon.
Chapter 3
46
Results and discussion
The two major carbon compounds found in grass silage, glucose and lactate (Krotscheck et
al., 2004), were investigated as substrates, individually and as mixtures, for the production
of L-lysine. To evaluate the potential of these two carbon sources, elementary mode
analysis was carried out using the metabolic network depicted in Figure S 3.1 and Table S
3.1 of the supplementary material. The resulting carbon yields for biomass and L-lysine on
pure glucose, pure lactate and also on mixtures of glucose and lactate are depicted in
Figure 3.2.
0 500 1000 1500 2000 2500 3000 3500 4000 4500 50000
0.2
0.4
0.6
0.8
1
YLy
s/S [
C-m
ol C
-mol
-1]
0 500 1000 1500 2000 2500 3000 3500 4000 4500 50000
0.2
0.4
0.6
0.8
1
YB
M/S
[C
-mol
C-m
ol-1
]
Figure 3.2: L-lysine and biomass yielding modes on mixtures of lactate and glucose
sorted with carbon yield of L-lysine with first priority and biomass with second
priority. A – carbon yield of L-lysine on glucose and lactate; B – carbon yield of
biomass on glucose and lactate.
Chapter 3
47
Many of these modes are so called extreme modes, producing either biomass or L-lysine
exclusively, being displayed on the axes. In addition, numerous modes show simultaneous
production of biomass and L-lysine. In Figure 3.2 all biomass and L-lysine yielding modes
are plotted and sorted with respect to L-lysine yield as first priority and biomass yield as
second priority. Maximal L-lysine carbon yield, YLys/s, on lactate was 75%. Interestingly
this mode used gluconeogenesis and NADPH production exclusively via glucose-6-
phosphate dehydrogenase (R11) (Table S 3.2 of supplementary material). A best yielding
mode excluding R11 is also listed there and resulted in 72.2%. In this mode NADPH was
mostly produced by malic enzyme (R48) and to a lesser extent by isocitrate dehydrogenase
(R32). Highest L-lysine carbon yield on glucose was 85.7% (Table S 3.3 of supplementary
material), and this is larger than literature values (Kjeldsen and Nielsen, 2009; Melzer et
al., 2009; Shinfuku et al., 2009). This mode was, however, not using PTS (R8) but only
glucokinase (R60) that is usually not considered active in the wild type and produced equal
amounts of NADPH by R11 and R48. It is a pure anaerobic mode where all electrons and
ATP are completely balanced. Best mode after exclusion of R60 resulted still in 83.3%.
Additional exclusion of malic enzyme (R48) gave a highest yield as described in the
literature with 75%. For calculating the maximum yielding modes on glucose Kjeldsen and
Nielsen (Kjeldsen and Nielsen, 2009) excluded malic enzyme activity achieving also a L-
lysine yield of 75%. All the identified high yielding modes were using the dehydrogenase
branch of L-lysine biosynthesis. The highest L-lysine carbon yield of an elementary flux
mode with simultaneous consumption of glucose and lactate was 84.8% (Table S 3.4 of
supplementary material). Regarding the modes with simultaneous biomass and L-lysine
biosynthesis, the maximal, theoretical carbon yield for L-lysine with biomass production
were 71% on glucose, 53% on lactate and 53% with simultaneous utilization of glucose
and lactate with associated biomass carbon yields of 12%, 18% and 19%.
We used the identified elementary modes to get possible hints for useful targets for the
engineering of C. glutamicum. The method presented by Melzer et al. (2009) plotting
individual rates versus L-lysine production showed interesting correlations as depicted in
the supplementary Figure S 3.2 but the correlation was rather poor with our network
including glucokinase and malic enzyme activity.
Therefore we ranked the modes in different way using the objective function
Chapter 3
48
=
⎛ ⎞= ⎜ ⎟⎜ ⎟+⎝ ⎠∑
2
1
ni
i Lys BM
rOFr r
(1)
Where ri is any rate of the n reactions contained in the network listed in Table S 3.1
(supplementary material) and rLys and rBM are the rates of production of L-lysine (R61) and
biomass (R63) in that mode. By taking the square value, modes with multiple passages,
e.g. in a cycle, are thus penalized. Modes with lowest objective function, OF, use most
likely less enzymes and are thus expected to be more efficient than those with a high value
of OF. This prediction is, however, not taking into account the unknown effect of different
specific activities of all involved enzymes that may vary considerably. The resulting
ranked values depicted in Figure S 3.3 show that low values of OF are generally
correlating with high yields, both of L-lysine and biomass. In a next step we tried to
identify possible targets for overexpression or deletion using the function shown below
mod
, /1
n ei
i j Lys Sj Lys BM j
rLoad Yr r=
⎛ ⎞= ⎜ ⎟⎜ ⎟+⎝ ⎠∑ (2)
This function cumulates the rates of each reaction relative to the L-lysine and biomass
production rates weighted with the L-lysine yield on both substrates of that mode. In
Figure 3.3 it can be seen that for pure growth on lactate the most promising targets seem
pyruvate carboxylase (R41) and malic enzyme (R48) since the cumulative function as
defined in Equation 2 is initially increasing most significantly.
Chapter 3
49
0 2000
10
20
Lo
ad
-R4
1
Mode0 200
0
10
20
Lo
ad
-R4
3
Mode0 200
0
10
20
Lo
ad
-R4
8
Mode
0 2000
10
20
Lo
ad
-R4
2
Mode0 200
0
10
20
Lo
ad
-R1
0
Mode0 200
0
10
20
Lo
ad
-R1
1
Mode
0 200
0
10
20
Lo
ad
-R2
6
Mode0 200
0
10
20
Lo
ad
-R3
2
Mode0 200
0
10
20
Lo
ad
-R3
9
Mode
Figure 3.3: Elementary modes ranked with increasing objective function, OF, and
selected cumulative enzyme loadings as defined in Eq. (2) for modes with pure growth
on lactate. Reaction numbers are defined in Supporting information, Figure S 3.1 and
Table S 3.1.
Interestingly, the second important group of genes is related to gluconeogenesis, e.g.
phosphoenolpyruvate carboxykinase (R43) and glucose-6-phosphate isomerase (R10) and
the pentose phosphate cycle, e.g. glucose-6-phosphate dehydrogenase (R11). Taking
lactate (Figure 3.3) and all substrate combinations (supplementary Figure S 3.4 to S 3.6),
pyruvate carboxylase is the most significant enzyme and malic enzyme, gluconeogenic and
pentose phosphate pathway enzymes follow. These are therefore all possible targets for
overexpression. For pure glucose consumption (supplementary Figure S 3.5) pyruvate
carboxylase, malic enzyme and pentose phosphate pathway enzymes seem important while
PEP carboxykinase seems less important and glucose-6-phosphate isomerase (R10) is
nearly not required at all. In the analysis comprising all possibilities of substrate
Chapter 3
50
consumption (supplementary Figure S 3.4) again pyruvate carboxylase and glucose-6-
phosphate dehydrogenase seem very important but glucose-6-phosphate isomerase (R10)
seems also relevant working in the reverse direction thus contributing to increased NADPH
production. From this analysis we expected a reduction of L-lysine formation when
glucose-6-phosphate isomerase was not operating. We checked the relevance of glucose-6-
phosphate isomerase for the growth and L-lysine production on glucose-lactate mixtures
by deleting it. Glyoxylate pathway is also a possible option for the production of L-lysine
on lactate, but did not seem important using elementary flux mode analysis.
As a starting point, however, the most obvious target was constitutive overexpression of
the D-lactate dehydrogenase (dld). C. glutamicum has two genes for lactate oxidation,
encoding L–lactate dehydrogenase (lldD) and D–lactate dehydrogenase (dld). L-lactate
dehydrogenase is inducible, D-lactate dehydrogenase is constitutively expressed at a very
low rate (Cocaign-Bousquet and Lindley, 1995). C. glutamicum was engineered to improve
growth on L and D- lactate mixtures, being able to metabolize the D-lactate fraction in a
more efficient way.
Overexpression of the dld, pyc and malE genes.
The D-lactate dehydrogenase exhibited a low activity of 0.08 µmol (mg prot)-1 min-1 in the
parental strain C. glutamicum lysCfbr, together with poor growth and incomplete lactate
consumption (Table 3.3, supplementary Figure S 3.7). The D-lactate fraction was not
metabolized. From the initial 10 g/L racemic lactate 4.3 g/L D- lactate remained until the
end of cultivation as determined using an enzymatic test. In order to increase the dld
expression, the native promoter of the gene was replaced by the strong promoter Psod from
the gene sod of C. glutamicum, encoding the superoxide dismutase. The mutant strain C.
glutamicum lysCfbrdldPsod showed increased dld activity of 0.42 ± 0.09 µmol (mg prot)-1
min-1. The same promoter was used for the overexpression of pyruvate carboxylase and
malic enzyme. The specific activity of the malic enzyme was 0.21 ± 0.11 µmol (mg prot)-1
min-1 in the parental strain and 0.96 ± 0.21 µmol (mg prot)-1 min-1 in the mutant strain C.
glutamicum lysCfbr dldPsod pycPsod malEPsod. All genetic modifications were verified by
Chapter 3
51
PCR, proving the integration of the desired constructs into the genome. Enzyme data
proved that the use of a stronger promoter instead of the native one allowed an
overexpression of the targeted genes.
Table 3.3: Growth characteristics of mutants of C. glutamicum ATTC created in this
study. µ - specific growth rate; YX/S – biomass yield on substrate given as g cell dry
weight per C-mol of substrates; YLys/S – yield of L-lysine on substrates given in carbon
moles per carbon mole.
Strain µ [ h-1] YX/S [g CDW/C-mol S ]
YLys/S [C-mol Lys/C-mol S ]
Lactate Gluc/Lac Lactate Gluc/Lac Lactate Gluc/Lac
lysCfbr 0.22 ± 0.01 0.13 ± 0.01 0.18 ± 0.06 0.21 ± 0.07 0 0
lysCfbr
dldPsod 0.34 ± 0.01 0.28 ± 0.02 0.21 ± 0.04 0.37 ± 0.02 0 0.05 ± 0.01
lysCfbr dld pyc Psod
0.38 ± 0.02 0.39 ± 0.01 0.29 ± 0.04 0.43 ± 0.07 0 0.08 ± 0.02
lysCfbr dld malEPsod
0.36 ± 0.01 0.28 ± 0.0 0.30 ± 0.02 0.38 ± 0.05 0.07 ± 0.01 0.11 ± 0.03
lysCfbr dld pyc
malEPsod 0.29 ± 0.0 0.22 ± 0.02 0.23 ± 0.02 0.39 ± 0.07 0.09 ± 0.04 0.15 ± 0.01
lysCfbr dld pyc
malEPsod
∆pgi
- 0.26 ± 0.02 - 0.33 ± 0.06 - 0.13 ± 0.02
Chapter 3
52
Effects on growth and L-lysine production on lactate.
The parental strain C. glutamicum lysCfbr was cultivated on minimal medium with D,L-
lactate as sole carbon and energy source (Figure S 3.7 of the supplementary material). The
growth rate (µ = 0.22 h-1) and the carbon yield of biomass of 18% were low whereby the
strain was not able to utilize D-lactate effectively (Table 3.3). The specific lactate uptake
rate calculated from data contained in Table 3.3 was 1.22 (C-mol S) (C-mol X)-1 h-1. As
stated above, 4.3 g/l of D-lactate remained at the end of this culture from the original 10 g/l
racemic lactate. The mutant C. glutamicum lysCfbrdldPsod showed increased levels of D-
lactate dehydrogenase activity and faster growth (µ = 0.34 h-1) due to complete lactate
consumption as well as an 30% increased specific uptake rate for D,L-lactate of qslac =
1.62 (C-mol S) (C-mol X)-1 h-1 (calculated from Table 3.3). Additional overexpression of
pyruvate carboxylase further increased the growth rate to µ = 0.38 h-1. Enhancing the
anaplerotic activity by the pyc overexpression obviously increased the supply of biomass
precursors of the TCA cycle, glycolysis and pentose phosphate pathway as well as of
NADPH resulting in a higher carbon yield of biomass of 29% (Table 3.3). Compared to the
wild type strain the specific substrate uptake rate could be increased by 7% to qslac = 1.31
(C-mol S) (C-mol X)-1 h-1. Improved availability of precursor molecules and NADPH
obviously led to the higher carbon yield of biomass and growth rate. Overexpression of
malic enzyme in the C. glutamicum lysCfbrdldPsodpycPsodmalEPsod strain led to a decreased
growth rate of µ = 0.29 h-1 and biomass yield of 23%. The growth on lactate was
exponential (Figure 3.4A) and well balanced with biomass and L-lysine production
inversely proportional to lactate consumption (Figure 3.4B and 3.4C). This also goes along
with a substrate uptake rate of qslac = 1.26 (C-mol S) (C-mol X)-1 h-1, 3% larger than that of
the wild type strain, and a substantial shift in the distribution of the substrate carbon from
biomass towards L-lysine production with a carbon yield of 13% as described in more
detail later.
Chapter 3
53
0 2 4 6 80
2
4
6
8
10
0
50
100
150
200
0
2
4
6
8
CD
W [g
L-1]
Time [h]
Lac
tate
[mM
]
L
ysin
e [m
M]
0 50 100 150 200 250 300 350 4000
10
20
30
40
Lys
ine
[C-m
Mol
]
Lactate [C-mMol]
y = 40.2 - 0.094 X
0 50 100 150 200 250 300 350 4000
20
40
60
80
100
120
CD
W [C
-mM
ol]
Lactate [C-mMol]
y = 106.6 - 0.228 x
Figure 3.4: Growth and L-lysine production C. glutamicum ATCC 13032
lysCfbrdldPsodpycPsodmalEPsod on D,L-lactate as sole carbon source. (A) Concentration
profiles; (B, C) cell dry weight and L-lysine and lactate carbon concentrations.
Chapter 3
54
Effects on growth and L-lysine production on lactate-glucose mixtures.
When grown on minimal medium with a glucose-lactate mixture as carbon source, the
parental strain C. glutamicum lysCfbr had a growth rate of µ = 0.13 h-1 (Table 3.3). During
growth on a glucose-lactate mixture lactate was metabolized at a specific rate equal to the
case of pure lactate growth. Glucose was metabolized simultaneously but at a decreased
consumption rate compared to pure glucose consumption (data not shown). After complete
consumption of lactate, the glucose consumption rate increased back to that observed
during growth on pure glucose. Similar observations concerning other organic acids are
described in the literature (Georgi, 2006). The wild type strain had a poor D-lactate uptake
that was even decreased in the presence of glucose yielding a value of qslac/glu = 0.62 (C-
mol S) (C-mol X)-1 h-1. This is compatible with the slow growth of the C. glutamicum
lysCfbr on the glucose-lactate mixture. The C. glutamicum lysCfbrdldPsod strain showed a
more than 100% increase of growth rate compared to the wild type with µ = 0.28 h-1 (Table
3.3) and consumed lactate significantly faster with qslac/glu = 0.76 (C-mol S) (C-mol X)-1 h-
1. Overexpression of the pyc gene further increased the growth rate to µ = 0.39 h-1 with an
even higher specific substrate uptake rate of qslac/glu = 0.91 (C-mol S) (C-mol X)-1 h-1. This
strain also had the highest carbon yield on the lactate-glucose mixture of 43% (Table 3.3).
C. glutamicum lysCfbrdldPsodpycPsodmalEPsod strain showed slower growth with µ = 0.22 h-1
and lower substrate uptake qslac/glu = 0.56 (C-mol S) (C-mol X)-1 h-1. Carbon yield of
biomass on the mixed substrates was 39% (Figure 3.5 and Table 3.3).
Lactate and glucose consumption were proportional to each other until complete
consumption of lactate (Figure S 3.8 in supplementary material), interestingly with a nearly
double carbon consumption rate of lactate (dLac/dGlc=1.96 C-mol C-mol-1). Concurrent
with the complete consumption of lactate, glucose uptake rate increased about threefold as
can be seen from the slope in the plot of glucose versus biomass concentration in Figure S
3.8, where the slope equals the ratio of glucose and biomass rates. In the presence of
lactate YX/Glc=1.15 (C-mol X) (C-mol Glc)-1 and after consumption of lactate YX/Glc=0.42 (
C-mol X) (C-mol Glc)-1. Since the specific growth rate remained constant, the specific rate
of glucose uptake increased by about 170%.
Chapter 3
55
Effects on L-lysine overproduction.
Surprisingly, the parental strain C. glutamicum lysCfbr did not produce detectable amounts
of L-lysine neither on lactate nor lactate-glucose mixtures (Table 3.3). This is in contrast to
the growth of this strain on pure glucose where a L-lysine carbon yield of 8.7% was
obtained . During growth on lactate, the available NADPH and carbon were obviously
primarily used for biomass formation though the aspartokinase is feedback resistant for L-
lysine in this strain. A possible explanation of the absence of L-lysine secretion is the
strong competition for oxaloacetate and NADPH by growth and L-lysine overproduction.
This interpretation is supported by results obtained after overexpression of dld showing
much faster lactate uptake but no beneficial effect on L-lysine production with lactate as
sole carbon and energy source. Even overexpression of pyc that should increase the supply
of oxaloacetate did not result in secretion of L-lysine during cultivation on lactate probably
caused by a still limited supply of NADPH. On a lactate-glucose mixture the C.
glutamicum lysCfbrdldPsodpycPsod mutant with the additionally overexpressed pyruvate
carboxylase had a L-lysine carbon yield of 8% and the highest biomass carbon yield of
43% (Table 3.3). The overexpression of pyc apparently leads to higher oxaloacetate
formation, a key precursor for L-lysine production, thus increasing L-lysine carbon yield
from 5% to 8%. The specific L-lysine production rate was nearly doubled from 0.038 to
0.073 (C-mol Lys) (C-mol S)-1. Overexpression of malic enzyme in the C. glutamicum
lysCfbrdldPsodpycPsodmalEPsod increased the L-lysine yield on lactate from 0% to 9% and on
lactate-glucose mixture from 8% to 15% corresponding to an 86% increase. Similarly,
overexpressing malic enzyme in the C. glutamicum lysCfbrdldPsod background increased L-
lysine yield from 0% to 8.4% on lactate and from 5% to 11% on the lactate-glucose
mixture, a 120% increase. This is most likely due to better NADPH supply, a key
prerequisite for L-lysine biosynthesis. The specific production rates increased from 0.073
to 0.085 (C-mol Lys) (C-mol S)-1, corresponding to 16% increase, and from 0.038 to 0.081
(C-mol Lys) (C-mol S)-1 corresponding to a 113% increase (Table 3.3). Kinetics of this
strain with highest L-lysine yield is shown in Figures 3.4 and 3.5. The growth on lactate
was exponential and well balanced with biomass and L-lysine production inversely
proportional to lactate consumption (Figure 3.4B and 3.4C). Interestingly biomass and L-
lysine production remained constant even after complete exhaustion of lactate (Figure S
3.8 of supplementary material).
Chapter 3
56
0 2 4 6 8 10 120
10
20
30
40
50
60
70
80
0
10
20
30
40 L
acta
te [m
M]
Time [h]
Glu
cose
[mM
]
0 2 4 6 8 100
2
4
6
0
2
4
6
8
10
12
14
16
CD
W [g
L-1]
Time [h]
Lys
ine
[mM
]
0 50 100 150 200 250 300 350 400 4500
20
40
60
80
100
120
140
160
180
CD
W [C
-mM
ol]
Substrate [C-mMol]
y = 195.2 - 0.395 x
0 50 100 150 200 250 300 350 400 4500
10
20
30
40
50
60
70
80
Lys
ine
[C-m
Mol
]
Substrate [C-mMol]
y = 67.63 - 0.145 X
Figure 3.5: Growth and L-lysine production C. glutamicum ATCC 13032
lysCfbrdldPsodpycPsodmalEPsod on a mixture of D,L-lactate and glucose. (A) Glucose and
lactate concentrations; (B) cell dry weight (CDW) and L-lysine; (C, D) CDW and L-
lysine and total substrate carbon concentrations.
Deletion of the phosphoglucose isomerase (pgi).
The elementary modes analysis predicted a substantial promotion of L-lysine yields on
lactate by activation of gluconeogenesis and pentose phosphate but less with lactate-
glucose mixtures (supplementary Figures S 3.4 to S 3.6). However, it was of great interest
how significant the contribution of these pathways to L-lysine production was in a real
strain producing L-lysine on glucose-lactate mixtures. To study the importance of
gluconeogenesis and particularly the supply of NADPH by the oxidative branch of the
pentose phosphate pathway, glucose-6-phosphate isomerase was deleted in the C.
glutamicum lysCfbrdldPsodpycPsodmalEPsod background. This deletion is channeling the
whole glucose carbon flux through the PPP, with no split to glycolysis resulting in a
production of two moles of NADPH per mole of glucose consumed. Deletion of pgi would,
Chapter 3
57
however, not allow further carbon entering the oxidative part of the pentose phosphate
pathway. Since improvement of the L-lysine yield is often linked to an increased flux
through the PPP associated with high yield of NADPH (Wittmann and Heinzle, 2002) the
glucose-6-phosphate isomerase seemed to be an interesting target to study the contribution
of different pathways to the observed increased L-lysine production (Marx et al., 2003).
The experimental results listed in Table 3.3 confirmed that glucose-6-phosphate isomerase
is important for L-lysine overproduction on glucose-lactate mixtures and that its deletion is
not beneficial for further improving the L-lysine yield. While the deletion of the pgi in the
C. glutamicum lysCfbrdldPsodpycPsodmalEPsod strain increased the specific growth rate to
0.26 h-1, the L-lysine yield was reduced from 15% to 13%. Interestingly, however, the
specific production rate of L-lysine increased to an observed maximum of all conditions
studied here, i.e. 0.102 (C-mol Lys) (C-mol X)-1.
Chapter 3
58
Concluding remarks
In the present work, we have investigated the growth and L-lysine production of mutants
of C. glutamicum ATCC 13032 lysCfbr, serving as wild type strain in this study, on racemic
mixtures of lactate as well as on mixtures of lactate and glucose. The growth on racemic
lactate was significantly improved by the overexpression of dld both with respect to
growth rate (55% increase) as well as biomass yield (17% increase). The growth rate
increase is more than could be expected from a simple duplication of lactate supply to the
metabolic network via pyruvate. Additional overexpression of pyc led to a further 12%
increase of growth rate and a substantial 38% increase in biomass yield. Alternatively also
the overexpression of malic enzyme in the C. glutamicum lysCfbr dldPsod background led to
a 6% increase in growth rate and 43% in biomass yield. The results in Table 3.3 show that
for L-lysine production on lactate overexpression of malic enzyme is essential. The mutant
C. glutamicum lysCfbrdldPsod showed increased levels of D- lactate dehydrogenase activity
of 0.42 µmol (mg prot)-1 min-1. Elementary mode analysis combined with the objective
function of equation 1 and cumulative enzyme load calculated using equation 2 suggested
pyruvate carboxylase as the most significant enzyme followed by malic enzyme but also
gluconeogenic and pentose phosphate pathway enzymes. The overexpression of malic
enzyme turned out to be essential for L-lysine production on lactate whereas the
overexpression of dld alone did already result in 5% carbon yield in L-lysine. The parental
strain gave a L-lysine yield of 8.7% on glucose whereas on the lactate-glucose mixture L-
lysine was not produced at all. The best L-lysine yields were obtained in the C. glutamicum
lysCfbrdldPsodpycPsodmalEPsod strain, 9% carbon yield on lactate and 15% on a mixture of
lactate and glucose. The predicted decrease of L-lysine yields caused by the pgi deletion
could be experimentally confirmed. It seems that cyclic operation of the pentose phosphate
pathway is significantly contributing to L-lysine production via an increased supply of
NADPH as was previously observed for glucose, fructose and sucrose (Georgi et al., 2005;
Kiefer et al., 2004; Wittmann et al., 2004).
All together we conclude that elementary flux mode analysis combined with useful
objective functions permits a very useful prediction of potential targets for increasing L-
lysine production by overexpression of selected genes. This may also be applied
successfully in other cases, possibly combined with other useful methods as suggested by
Melzer et al. (Melzer et al., 2009) but also the recently published method of Park et al. that
Chapter 3
59
incorporated additional constraints of genomic context also of related organisms into target
identification (Park et al., 2010). Our computational and experimental results also point to
further potentially useful and scientifically interesting modifications, especially for
enhancing gluconeogenesis and leading to a stronger flux through the PPP, e.g.
overexpression of fructose 1,6-bisphosphatase (Georgi et al., 2005; Kiefer et al., 2004;
Wittmann et al., 2004). Other interesting targets for further improvement of the constructed
strains taken from the literature are the down – regulation of the TCA cycle by modifying
start codons of TCA cycle genes to reduce their expression (Becker et al., 2009) or
introducing a mutation in the malate quinone oxidoreductase (Ikeda et al., 2006). These
modifications would lead to a decreased TCA activity thus avoiding unnecessary carbon
loss. Application of genetically modified strains on real silage juice would require
additional modifications to use other substrates contained in the juice, e.g. xylose.
Acknowledgements. We thank the Fachagentur für Nachwachsende Rohstoffe (FNR) that
supported this work. We thank Prof. Roland Ulber of the University of Kaiserslautern for
successful coordination of the joint project. We are very grateful for the most valuable
support by Michel Fritz concerning instrumental analysis.
Conflict of interest statement. The authors have declared no conflict of interest.
Chapter 4
60
Chapter 4
Production of L-lysine on different silage
juices using genetically engineered
C. glutamicum*
*Published as: Andreas Neuner, Ines Wagner, Tim Sieker, Roland Ulber, Konstantin Schneider, Susanne Peifer and Elmar Heinzle (2012): Production of L-lysine on different silage juices using genetically engineered Corynebacterium glutamicum, Journa of Biotechnology, In Press.
Chapter 4
61
Abstract
Corynebacterium glutamicum, the best established industrial producer organism for L-
lysine was genetically modified to allow the production of L-lysine on grass and corn
silages. The resulting strain C. glutamicum lysCfbr dldPsod pycPsod malEPsod fbpPsod gapXPsod
was based on earlier work (Neuner and Heinzle, 2011). That mutant carries a point
mutation in the aspartokinase (lysC) regulatory subunit gene as well as overexpression of
D-lactate dehydrogenase (dld), pyruvate carboxylase (pyc) and malic enzyme (malE) using
the strong Psod promoter. Here, we additionally overexpressed fructose 1,6-bisphosphatase
(fbp) and glyceraldehyde 3-phosphate dehydrogenase (gapX) using the same promoter. The
resulting strain grew readily on grass and corn silages with a specific growth rate of 0.35 h-
1 and L-lysine carbon yields of approximately 90 C-mmol (C-mol)-1. L-lysine yields were
hardly affected by oxygen limitation whereas linear growth was observed under oxygen
limiting conditions. Overall, this strain seems very robust with respect to the composition
of silage utilizing all quantified low molecular weight substrates, e.g. lactate, glucose,
fructose, maltose, quinate, fumarate, glutamate, leucine, isoleucine and alanine.
Chapter 4
62
Introduction
European green biorefinery concepts are usually based on silage guaranteeing a constant
supply over the year (Hanegraaf et al., 1998; McDonald et al., 1991). Ensiling is the
process of preserving the wet plant material applying anaerobic conditions, either in a
storage silo or wrapped in plastic. After harvest, the green plant material is chopped and
left to wither up to a dry mass content of approximately 30% followed by the removal of
oxygen by compression assuring anaerobic conditions. Lactic acid bacteria convert a large
fraction of water soluble carbohydrates into lactic acid. The low pH (3-4) and the
anaerobic conditions prevent coliform bacteria and clostridia from spoiling the crop
(McDonald et al., 1991). On a large scale, silage is either used for the production of biogas
or as animal feed (Kromus et al., 2004). In the case of biogas production it seems
interesting to convert part of the easily accessible compounds into more valuable products,
e.g. amino acids, and use the remainder for biogas production. We aimed at the production
of L-lysine using Corynebacterium glutamicum (Eggeling and Bott, 2005). Since C.
glutamicum is a GRAS organism, fermentation broth containing L-lysine and C.
glutamicum could be used as feed supplement for monogastric animals in close-by chicken
and pig farming (Leuchtenberger et al., 2005). A second alternative might be a direct
combination of L-lysine fermentation of silage to enrich silage with L-lysine. Only small
increases in L-lysine content improves the biological value measured by the protein
efficiency ratio significantly (Belitz, 2001). This makes L-lysine the most used amino acid
in animal feed supplementation (Belitz, 2001) improving the nitrogen uptake, enhancing
growth (Leclercq, 1998) and decreasing the release of nitrogen into the environment.
Dominating compounds in sugar beet and corn silages are sucrose and starch, respectively,
whereas grass silage does not have a major component. Lactic acid itself is a valuable
intermediate for the manufacture of biodegradable polymers (Södergard, 2002) but
recovery costs from silage are very high (Datta et al., 1995). Therefore, we try to use it as
an carbon source for L-lysine production (Neuner and Heinzle, 2011).
Chapter 4
63
C. glutamicum, a gram positive, non-pathogenic soil bacterium grows aerobically on
various carbohydrates and organic acids as carbon sources (Liebl et al., 1991) but only
poorly on racemic lactate (Neuner and Heinzle, 2011). Applying metabolic network
analysis using elementary modes (Kjeldsen and Nielsen, 2009; Krömer et al., 2006; Melzer
et al., 2009; Neuner and Heinzle, 2011; Schuster et al., 1999) to guide genetic engineering,
C. glutamicum was engineered to grow on racemic lactate and on carbohydrates and
mixtures thereof and at the same time producing L-lysine (Neuner and Heinzle, 2011).
In this work, the L-lysine producing strain C. glutamicum lysCfbr dldPsod pycPsodmalEPsod
(Neuner and Heinzle, 2011) was further genetically modified and tested for efficient L-
lysine production on different silage juices. We focused on the three main carbon sources,
glucose, fructose and lactate. However, other sugars and organic acids existing in silage
juices (Wiseman and Irvin, 1957) were also analyzed. Besides earlier identified and tested
modifications (Neuner and Heinzle, 2011) we studied two additional targets for
overexpression that were predicted by the same model and method but using lactate,
glucose and fructose as substrates. The gluconeogenic enzyme fructose 1,6-bisphosphatase
(fbp) was already earlier found being beneficial for L-lysine production on carbohydrates
(Becker et al., 2005; Becker et al., 2011; Rittmann et al., 2003) and therefore was
constitutively overexpressed. Furthermore, glyceraldehyde 3-phosphate dehydrogenase
that was also predicted as beneficial for L-lysine production was overexpressed. We
studied growth and L-lysine production of the resulting strain on different silage juices and
under different cultivation conditions and found substantial L-lysine production even under
oxygen limitation as was already observed previously (Ensari and Lim, 2003).
Chapter 4
64
Material and methods
Pretreatment of silage juice
The silages we used were provided by the Lehr- und Versuchsanstalt für Viehhaltung,
Hofgut Neumühle (Münchweiler an der Alsenz, Germany). The silage juices were obtained
by using a HP2H tincture press (Fischer Maschinenfabrik GmbH, Neuss, Germany). We
applied filtration with stericups (Merck Millipore, Darmstadt, Germany) and
pasteurization, using different temperatures for different periods of time. When using
filtration with 0.22 µm pore size, the silage juices were centrifuged previously at maximum
speed (16,000 × g) for 10 minutes to remove the majority of the particles from the
suspension. The pH was adjusted to a value of 7 using 30% ammonia solution or 2 N
NaOH. Heat pre-treatment tests were only made on grass silage. After the heat pre-
treatment, silage juice was plated out on LB and CM agar plates and incubated at 30°C and
37°C for one week. Colony formation was observed. Subsequently, the treated juice was
used for growth experiments with C. glutamicum mutants.
Strains, plasmids and recombinant DNA methods
All mutants were designed on basis of the L-lysine producing strain C. glutamicum ATCC
13032 lysCfbr (BASF AG, Ludwigshafen, Germany) with deregulated L-lysine biosynthesis
(allelic replacement of the lysC gene with a lysCT311I gene) (Kim et al., 2006).
Overexpression of the genes fbp and gapX in C. glutamicum was achieved by cloning the
open reading frame (ORF) of the mentioned genes under the control of the strong
constitutive promoter of the sod gene, encoding superoxide dismutase (NCgl2826).
E. coli DH5α was used for the amplification of the genetic constructs. Application of the
methylation pattern of C. glutamicum to the genetic constructs was performed using E. coli
NM522, containing the pTc plasmid as an expression vector for the DNA –
Chapter 4
65
methyltransferase of C. glutamicum. The integrative plasmid pClik int sacB, carrying a
kanamycin resistance and the sacB gene as selective markers was used for introducing the
genetic modifications. Transformation of the organism with the plasmid and selection for
kanamycin resistance yielded transformants with genome integrated plasmid DNA.
Integration of the plasmid DNA occurred via a single crossover homologous
recombination. The second recombination was detected and selected via the sacB positive
selection system (Jäger et al., 1992). Sucrose resistant, kanamycin sensitive clones were
tested for the presence of the mutation by PCR. In addition, sequencing of the resulting
PCR product was performed (GATC, Konstanz, Germany). Detailed information about all
bacterial strains and plasmids used in this study, their relevant characteristics and their
sources are listed in Table 4.1. Used primers are listed in Table 4.2.
Table 4.1: Molecular and biological tools and strains constructed starting from C.
glutamicum ATCC 13032.
Strain Modifications Reference
C. glutamicum lysCfbr Exchange T311I in the lysC gene BASF C. glutamicum
lysCfbrdldPsodpycPsodmalEPsod
lysC T311I + Exchange of the natural promoter of the dld,
pyc and malE genes by the promoter of the sod gene
Neuner and Heinzle, 2011
C. glutamicum
lysCfbrdldPsodpycPsodmalEPsodfbpPsod
lysC T311I + Exchange of the natural promoter of the dld,
pyc, malE, fbp genes by the promoter of the sod gene
This work
C. glutamicum
lysCfbrdldPsodpycPsodmalEPsodfbpPsodgap
XPsod
Referred to as C. glutamicum strain SL.
lysC T311I + Exchange of the natural promoter of the dld,
pyc, malE, fbp, gapX genes by the promoter of the sod gene
This work
E.coli DH5α
F- endA1, hsdR17 (vk- mk+) supE44, thi-I λ- recAI gyrA96
rel A1, ∆ (lac ZYA-argF)U169 F80d lacZ ∆M15
(Hanahan, 1983)
E.coli NM522
supE thi-1 ∆(lac-proAB)∆(mcrB-hsdSM) 5(rK- mK+) [F’
proAB laclq Z∆M15]
Stratagene
Plasmid
pClik int sacB vector for integrative, allelic replacement by homologuous
recombination, nonreplicative in C. glutamicum, KanR,
sacB
BASF
Chapter 4
66
Table 4.2: Primer sequences used for verification of the mutant strains and the PCR
fragment size of the wild type and mutant alleles.
Genetic modification
Primer sequence PCR fragment size
PsodgapX
Fw: 5’- ATCGACGCGTTCGCAGCCGGCGGCCTTTCAACCTCCG - 3‘
Rw: 5’ - CGATCTCGAGCGCCAGCGGCCGGTGTTGTCTACCACGACG - 3‘
WT: 1218 bp
Psoddld: 1410 bp
Psodfbp Fw: 5’- AGTTGCATGATCAGTCATTGCGCGCGCTTCC -3’
Rw: 5’- AGTCTGTCCACCAGCTGTCCAAGCTGCAGGAATAC - 3’ WT: 1377 bp
Psodpyc: 1569 bp
For the construction and purification of plasmid DNA standard protocols were applied.
Chromosomal DNA from C. glutamicum was obtained using the Instant Bacteria DNA Kit
(Analytic Jena, Jena, Germany). Plasmids from C. glutamicum were isolated using the
HiSpeed® Midi Kit (Quiagen, Hilden, Germany). Plasmids from E. coli were recovered
using the GFXTM PCR DNA and Gel Band Purification Kit (GE Healthcare, Munich,
Germany). Oligonucleotide synthesis was carried out by Sigma (Munich, Germany). All
PCR reactions were done in a TGradient-Cycler (Whatman - Biometra®, Goettingen,
Germany), with FidelyTaq (Fermentas, Mannheim, Germany) or Jumpstart RedTaq Mix
(Sigma, Munich, Germany) using 10 mM dNTP Mix and the 1kb DNA O’GeneRulerTM
(Fermentas, Mannheim, Germany). Ligations were performed with Fast-LinkTM DNA
Ligation Kit (Epicentre Biotechnologies, Madison, USA). Restriction enzymes and their
buffers were obtained from Fermentas (Mannheim, Germany) and used as recommended
by the manufacturers. E. coli was transformed by heat shock. C. glutamicum was
transformed by electroporation (Tauch et al., 2002b; van der Rest, 1999). Strain
verification was done by DNA sequencing, GATC, Konstanz, Germany.
Media
For the bioreactor cultivations, filtered silage juice was diluted with water (1:4; v:v) and
then directly used as fermentation broth. For sterility tests we used LB (5 g/L yeast extract,
10 g/L tryptone, 10 g/L NaCl, 20 g/L agar, pH 7) and CM (10 g/L glucose, 2.5 g/L NaCl, 2
g/L urea, 5 g/L yeast extract, 5 g/L peptone, 20 g/L casaminoacids, 20 g/L agar, pH 7) agar
plates. Genetic manipulations of C. glutamicum have been performed using different
complex media containing 10 g L-1 glucose, 5 g L-1 yeast extract, 5 g L-1 beef extract, 5 g
Chapter 4
67
L-1 polypeptone, 20 g L-1 casaminoacids, 2.5 g L-1 NaCl and 2 g L -1 urea. The selection of
positive mutants after the first recombination was done on CMKan agar plates,containing 20
mg/ml kanamycine. The selection after the second recombination on CMSac was made on
agar plates containing 100 g L-1 sucrose. For agar plates, 20 g L-1 agar were added. After
electroporation, cells were incubated in BHIS medium containing 37 g L-1 BHI and 250 ml
2M sorbitol solution for recovery. All other chemicals and reagents of analytical grade
were purchased from Sigma-Aldrich, Merck (Darmstadt, Germany), Fluka (Buchs,
Switzerland) or Roth (Karlsruhe, Germany).
Cell disruption
Cells were harvested by centrifugation (5 min, 6,500 x g, 4°C). 100 mg of glass beads (Ø
0.25 mm) and 1 mL deionized purified water were added to the pellet. Disruption was
performed using a bead mill (Retsch, MM301, Haan, Germany) at 30 Hz for 35 seconds.
Cell debris was removed by centrifugation.
Fructose 1,6-bisphosphatase activity
The in vitro FBPase activity in cell extracts of C. glutamicum was measured in a coupled
spectrophotometric assay containing 100 mM Tris/HCl, 5 mM MnCl2, 0.5 mM NADP+, 2
U/mL of phosphoglucoisomerase, 1 U/mL of glucose 6-phosphate dehydrogenase, pH 7.7,
30°C, 50 µl cell extract for the reaction mix and 100 mM fructose 1,6-bisphosphate, 100
mM Tris/HCl, pH 7.7 for the substrate solution. Temperature optimum was determined to
be 30°C. The assay was started by adding 50 µl substrate solution to 950 µl reaction mix.
Fructose 6-phosphate formed by the reaction of FBPase was converted to glucose 6-
phosphate and subsequently to 6-phosphogluconate by phosphoglucoisomerase and
glucose 6-phosphate dehydrogenase. The formation of NADPH (ε340 nm = 6.22 mM–1 cm–1)
was monitored at 340 nm. Negative controls were carried out without fructose 1,6-
bisphosphate or without cell extract, respectively. All measurements were performed in
triplicate.
Chapter 4
68
Glyceraldehyde 3-phosphate dehydrogenase activity
The in vitro activity of glyceraldehyde 3-phosphate dehydrogenase was determined
according to a published method (Crow and Wittenberger, 1979). The reaction mixture
contained 1 mM NAD+, 5 mM sodium arsenate, 5 mM cysteine/HCl, 125 mM
triethanolamine and cell extract, pH 7.5. The substrate mixture contained 2 mM D,L-
glyceraldehyde 3-phosphate. Glyceraldehyde 3-phosphate dehydrogenase was assayed
after 1 h incubation in triethanolamine/HCl buffer, containing cysteine/HCl to reactivate
the enzyme by reducing sulfhydryl groups. The assay was initiated by the addition of
glyceraldehyde 3-phosphate. The assay is based on the coupling of enzyme activity to the
consumption or production of NADH, monitored at 340 nm in a spectrophotometer (ε340 nm
= 6.22 mM–1 cm–1). All measurements were performed in triplicate. Negative controls were
carried out without substrate or without cell extract, respectively. All measurements were
performed in triplicate. Protein concentration was determined by the method of Bradford
(Bradford, 1976) with bovine serum albumin as a standard.
Shake flask cultivation
Shake flask cultivations were performed using undiluted, filtrated grass and corn silage
juice, respectively. The precultures and main cultures in shake flasks were performed using
the same medium, undiluted silage juice. Single colonies from agar plates were used as
inoculums for the first preculture, which was grown for 8 h at 30°C in a 100 mL baffled
shake flask with 10 mL medium on a rotary shaker (Multitron II, Infors AG, Bottmingen,
Switzerland) at 230 rpm. Cells were harvested by centrifugation (3 min, 6,000 rpm, 4°C,
Labofuge 400R, Heraeus, Hanau, Germany) and used as inoculum for the main culture.
Main cultures were performed in duplicate using 250 mL baffled shake flasks with 25 mL
medium. Cell concentration was determined photometrically (Novaspec®II, Pharmacia
Biotech, Little Chalfont, UK) at 660 nm.
Bioreactor cultivation
Pre-cultures and main cultures were performed in 1:4 (v:v) diluted silage juice, providing
all necessary macro- and micronutrients. Cultivations were performed in a 2.3 L bioreactor
Chapter 4
69
RK 01 – 23, (FairMenTech GmbH, Göttingen, Germany) with 1000 mL working volume
at 30°C, pH 7.0, 800 rpm, unless stated otherwise, The PEEK stirrer shaft diameter was 10
mm and the outer diameter ring carrying the blades 16 mm. The ring carried six impeller
blades of 10 mm height and 3 mm thickness. The total impeller diameters were 35 mm for
the small (impeller 1) and 52 mm for the large impeller (impeller 2). A three stage stirrer
was used. Dissolved oxygen was measured via a pO2 probe (Broadley –James Corporation,
Irvine, USA). Aeration rates were adjusted to meet the oxygen requirements of the strains
using a mass flow controller (Brooks Instruments, Veenedaal, Netherlands). From
measured gas flow rates and composition of aeration and exhausted gas (quadrupole mass
spectrometer Omnistar, Balzers AG, Vaduz, Liechtenstein) oxygen uptake (OUR) and CO2
production rates (CPR) as well as the respiratory quotient (RQ) were determined on line
(Heinzle et al., 1990). During cultivation the pH was adjusted with 2 N HCl and 2 N
NaOH.
Analysis of substrates and products
Cultivation samples were centrifuged (3 min, 13,000 x g, 4°C, Heraeus, Hanau, Germany)
and the supernatant was used for the analysis of substrates and products. Quantification of
sugars and organic acids in diluted supernatants was performed by high pressure liquid
chromatography (Kontron Instruments, Neufahrn, Germany) using a Aminex HPX-87H
column (Bio-Rad, Hercules, Calif.) at 60°C with 7 mM H2SO4 mobile phase and a flow
rate of 1 mL min-1. Detection was carried out using a refraction index detector (sugars) or a
UV detector (210 nm, organic acids). Quantification of amino acids was performed as
described by Krömer et al. (2005). Alternatively, enzyme kits (Boehringer Mannheim, R
Biopharm AG, Darmstadt, Germany) were used for the quantification of D,L-lactate (D,L-
lactate UV – test) and glucose / fructose (D-Glucose / D-fructose UV – test).
Chapter 4
70
Results and discussion
Pretreatment of silage juice
Heat pretreatment experiments were carried out using grass silage. 1 kg of silage yielded
approximately 550 mL silage juice, depending on the silage. pH was adjusted to a value of
about 7 using 30% ammonia solution or NaOH. Autoclaving the silage juice (20 min, 121
°C) obviously produced inhibitory or even toxic compounds, presumably by the Maillard
reaction. Since the process was designed for a potential future application in farming units
and biorefineries, we examined pasteurization to minimize spoilage. The pasteurization
parameters were tested by exposing the silage juice to different temperatures for different
periods of time. By determination of the growth rate and comparison with growth rates
obtained on filtrated juice, we elucidated the optimal time and temperature for the
pasteurization of silage juice. The results are depicted in Table 4.3. As shown in Table 4.3,
even short treatment of 5 min at temperatures of 100°C caused growth inhibition. Since
silage juice contains amino acids and sugars, excess heat causes the Maillard reaction to
occur. In an alkaline environment this reaction is accelerated as the amino groups are
deprotonated and therefore more nucleophile. The pH of silage juice was adjusted to a
value of 7, and we applied temperatures of over 80°C. This is a possible explanation for the
growth inhibition after extended exposure of the silage juice to this temperature. For the
determination of the specific growth rate we inoculated with a higher OD660 of 2 in cases
of non-sterile silage juice (80°C for 5 and 10 minutes). Therefore, the measured growth
rate can be attributed to the clearly predominant organism, C. glutamicum. The results of
Table 4.3 show that pasteurization at 80°C for 15 min is sufficient to avoid spoilage by
undesired microorganisms and at the same time to guarantee maximum growth. This kind
of pretreatment could be easily implemented on a larger scale using a counter-current heat
exchanger system.
Chapter 4
71
Table 4.3: Heat treatment of the grass silage juice with resulting sterility and growth
characteristics. Sterility was tested by plating treated silage on LA and CM agar
plates and incubation at 30 C and 37 C for seven days. Sterility: sterile – no colonies
on any plate; not sterile – colonies observed. Growth of C. glutamicum was tested in
shake flask cultures.
Temp / Time 5 min µ [h-1] 10 min µ [h-1] 15 min µ [h-1]
80°C
not sterile
growth
0.35±0.02
not sterile
growth
0.34±0.03
sterile
growth
0.36±0.01
100°C
not sterile
growth
0.26±0.01
sterile
no growth
-
sterile
no growth
-
120°C
sterile
no growth
-
sterile
no growth
-
sterile
no growth
-
Validation of the fbp and gapX overexpression
To validate the overexpression of the fbp and gapX genes, the in vitro specific enzyme
activities in the precedent strain, C. glutamicum lysCfbrdldPsodpycPsodmalEPsod were
compared to the specific activities in the C. glutamicum lysCfbrdldPsodpycPsodmalEPsodfbpPsod
and C. glutamicum lysCfb dldPsodpycPsodmalEPsodfbpPsodgapXPsod mutants. Results are
displayed in Table 4.4. The activity data show fourfold increased activity of the fructose
1,6-bisphosphatase and a nearly 2.5-fold increased activity of glyceraldehyde 3-
phosphatase in the new strain, proving a successful overexpression of the fbp and gapX
genes.
Chapter 4
72
Table 4.4: Enzyme activities of fructose 1,6-bisphosphatase, fbp, and glyceraldehyde
3-phosphate dehydrogenase, gapX, in various strains.
Enzyme Strain Enzyme activity
(mU/mg protein)
fbp
C. glutamicum lysCfbrdldPsodpycPsodmalEPsod
21.3 ± 0.4
C. glutamicum lysCfbrdldPsodpycPsodmalEPsodfbpPsod 86.1 ± 2.6
gapX C. glutamicum lysCfbrdldPsodpycPsodmalEPsodfbpPsod 568 ± 16
C. glutamicum lysCfbrdldPsodpycPsodmalEPsodfbpPsodgapXPsod 1318 ± 29
Influence of fbp and gapX overexpression on L-lysine production
C. glutamicum lysCfbrdldPsodpycPsodmalEPsod was able to produce L-lysine on grass silage
juice, with a L-lysine carbon yield of 3.2 ± 0.2%.The additional overexpression of the fbp
gene in the mutant C. glutamicum lysCfbrdldPsodpycPsodmalEPsodfbpPsod resulted in an almost
threefold increased L-lysine carbon yield of 9.0 ± 1.1%. This proves the importance of the
fbp gene regarding L-lysine production, especially when gluconeogenic substrates are
used. Former studies showed a L-lysine carbon yield increase of 30% during growth on
fructose minimal medium after the overexpression of the fructose 1,6-bisphosphatase
(Becker et al., 2005). Using silage with lactate as main carbon source and fructose in
considerable amounts, this gluconeogenic enzyme plays an even more important role,
increasing the L-lysine yield threefold. Additional overexpression of gapX slightly
improved the L-lysine carbon yield to 9.4 ± 0.2% that was however not statistically
significant.
Influence of the fbp and gapX overexpression on growth
Besides strongly affecting the L-lysine production, the overexpression of the fbp gene also
influenced the specific growth rate and biomass formation. Compared to the C. glutamicum
lysCfbrdldPsodpycPsodmalEPsod strain (µ = 0.38 h-1), the specific growth rate µ slightly
decreased in the C. glutamicum lysCfbrdldPsodpycPsodmalEPsodfbpPsod mutant (µ = 0.35 h-1).
Chapter 4
73
The C. glutamicum lysCfbrdldPsodpycPsodmalEPsodfbpPsod mutant exhibited a 20% higher
biomass carbon yield (44%) than the precedent strain (36%). A possible reason is an
increased flux through the pentose phosphate pathway, generating more NADPH that
boosted L-lysine production but also lead to an increased growth requiring NADPH as
well. Former studies described a comparable increase of biomass formation on fructose of
about 10% using a strain overexpressing fbp (Becker et al., 2005). Additional
overexpression of gapX resulted in an almost identical growth rate (µ = 0.36 h-1) but
reduced biomass carbon yield of 38%.
Influence of the fbp and gapX overexpression on byproduct formation
Especially on grass silage juice, containing a significant amount of fructose,
dihydroxyacetone (DHA) was a byproduct secreted by the C. glutamicum
lysCfbrdldPsodpycPsodmalEPsod mutant. Towards the end of the cultivation, the DHA reached
a final level of 7.1 ± 0.1 mM. Overexpression of the fructose 1,6-bisphosphatase in the C.
glutamicum lysCfbrdldPsodpycPsodmalEPsodfbpPsod mutant decreased the DHA level by almost
50%. Additional overexpression of glyceraldehyde 3-phosphate dehydrogenase further
decreased the DHA secretion to 1.9 ± 0.1mM. In corn silage, byproduct formation in form
of DHA was insignificant. The results showing the effect of the fbp and gapX
overexpression on DHA formation on grass silage juice are depicted in Table 4.5.
Table 4.5: Formation of dihydroxyacetone in various mutants.
Strain Dihydroxyacetone (mM)
C. glutamicum lysCfbrdldPsodpycPsodmalEPsod 7.1 ± 0.1
C. glutamicum lysCfbrdldPsodpycPsodmalEPsodfbpPsod 3.2 ± 0.2
C. glutamicum lysCfbrdldPsodpycPsodmalEPsodfbpPsodgapXPsod 1.9 ± 0.1
Chapter 4
74
Cultivation experiments
Bioreactor cultivation
In all cultivation experiments applying shake flasks and bioreactors profiles of major
carbohydrates, lactate, L-lysine and biomass were determined. Additional carbon sources
like the organic acids quinate and fumarate as well as the four predominant amino acids in
silage juice, i.e. glutamine, leucine, isoleucine and alanine were determined at the
beginning and at the end of each experiment since their amounts were significantly lower
(Table 4.6).
Table 4.6: Initial concentration of minor substrates measured. In all cultivations all
these substrates were completely consumed at the end of the respective experiment.
Concentration [C-mmol L-1]
Grass silage juice Corn silage juice
F1 F2 F3 F4 F5
Quinate 37.4 34.8 38.8 41.4 43.2
Fumarate 32.8 29.1 34.4 36.6 33.1
Glutamate 17.9 16.4 14.8 13.9 14.7
Leucine 14.9 15.5 16.1 12.8 13.3
Isoleucine 17.7 17.1 18.6 14.4 15.9
Alanine 18.3 19.9 17.6 12.9 14.1
The cultivation profile of C. glutamicum SL on grass silage juice is shown in Figure 4.1.
All major carbon sources were consumed simultaneously as depicted in Figure 4.1A and
depleted according to their initial concentrations.
Chapter 4
75
Fig. 4.1: Batch cultivation of L-lysine producing C. glutamicum lysCfbr dldPsod
pycPsodmalEPsodfbpPsodgapXPsod (C. glutamicum SL) on 1:4 diluted grass silage juice in
the bioreactor using impeller 1. (A) Concentrations of substrates fructose (Fru),
glucose (Glu) and lactate (Lac), of product L-lysine (Lys) and of biomass provided as
Chapter 4
76
cell dry weight (CDW). (B) Oxygen uptake rate (OUR) and carbon dioxide
production rate (CPR). (C) Optical density (OD) and dissolved oxygen concentration
(DO – full line) given in % air saturation and stirrer adjustment: 1 – 800 rpm, 2 –
1000 rpm, 3 – 1500 rpm, 4 – 2000 rpm, 5 – 1000 rpm. Dashed lines indicate identified
growth phases I–III.
Glucose was exhausted at about 12 h when oxygen uptake rate (OUR) and carbon dioxide
production rate (CPR) reached a distinct maximum value (Figure 4.1B). Simultaneously
dissolved oxygen concentration (DO) increased rapidly. Fructose was depleted at about 17
h and the respiration rates reached another less pronounced maximum. After 18 h of
cultivation, all carbon sources were completely consumed also indicated by another local
maximum of respiration rates. Although grass silage was diluted 1:4 with water, the
aeration capacity of the bioreactor applied was insufficient. At time points 1, 2, 3 and 4
depicted in Figure 4.1C the stirrer speed was stepwise increased up to the maximum
possible value of 2000 rpm. Nevertheless, DO fell below 5% air saturation after 13 h. After
depletion of fructose DO increased again. In this culture the growth in phase I was
exponential with two different rates, 0.16 h-1 until about 7 h and 0.39 h-1 until about 10 h.
L-lysine concentration increased steadily throughout the cultivation, reaching a final
concentration of 1.7 g L-1. The fermentation profile shows three distinct phases, mostly
dependent on oxygen availability. In the first phase (0 h to 11 h), characterized by
exponential growth under oxygen sufficient conditions, glucose, fructose and lactate were
consumed simultaneously. L–lysine was secreted and accumulated in the culture to
eventually reach 0.48 g L-1 when glucose was completely depleted. Characteristic for this
first phase are non-limiting oxygen concentrations and exponential growth. OUR and CPR
were increasing indicating a higher oxygen demand and CO2 production by the increasing
biomass. During the second phase (11 h to 18 h), oxygen limitation no longer allowed
exponential growth. During this phase biomass formation slowed down while L-lysine
secretion resulted in a concentration increase to 1.6 g L-1. During this oxygen limited phase
CPR was higher than OUR most probably due to oxygen limitation. The resulting average
RQ-value was about 1.3 between 7 h to 22 h cultivation with large noise before and after
that due to error amplification (Heinzle et al., 1990). This value can be explained by the
use of lactate and other organic acids as carbon and energy sources and the higher degree
of reductance of L-lysine compared to carbohydrates (Stephanopoulos et al., 1998). In the
Chapter 4
77
third phase (18 h to 28 h), substrate consumption was complete and biomass and L-lysine
reached final levels of 6.9 g L-1 and 1.7 g L-1, respectively, resulting in 0.24 g L-lysine (g
biomass)-1. While dissolved oxygen levels increased, OUR and CPR decreased below 5
mmol L-1 h-1 remaining essentially constant after 23 hours. No significant byproduct
formation could be observed in this phase. The produced amount of carbon dioxide was
461 mmol L-1. Additionally accounting for produced biomass and L-lysine, a total carbon
output of 810 mmol versus 743 mmol carbon input was determined. This means that the
origin of only 8.3% of the carbon detected in products is not clear. The difference is
actually within the experimental error of these measurements of around 5% each. The
detected overproduction of carbon containing products may originate from other
compounds originally contained in the silage that were not analytically detected but
converted to biomass, L-lysine or CO2.
Non-limiting conditions concerning dissolved oxygen (always above 60% DO) were
achieved by gassing pure oxygen and using larger impellers, thus increasing the specific
power input and therefore increasing oxygen transfer rate. The cultivation profile is
depicted in Figure S 4.1. However, only slight differences could be observed compared to
the oxygen limiting conditions. The L-lysine yield slightly increased to 0.28 g L-lysine (g
biomass)-1, and a lower RQ value of 1.15 was determined. In general, one can concluded
that growth and L-lysine production seem to be quite independent on the varying oxygen
supply in both cases and only minor changes could be observed, however, reaching a L-
lysine carbon yield of 9.4% indicates the potential of further improvements based on well-
known genetic modifications (Becker and Wittmann, 2012).
As depicted in Figure 4.2 cultivations on corn silage juice showed a very similar profile
concerning biomass and product formation as well as substrate uptake characteristics and
specific growth rate. In contrast to grass silage juice corn silage has a starch content of
about 300 g L-1 that cannot be readily metabolized by C. glutamicum (Seibold et al., 2006;
Tateno et al., 2007). After 5 hours, DO fell below 5% and exponential growth switched to
linear growth until the exhaustion of major monomeric carbon sources. After 13 hours, the
resulting concentration of L-lysine was 1.37 g L-1 and of biomass 6.1 g L-1 corresponding
to 0.23 g L-lysine (g biomass)-1, a value almost equal to the one of grass silage shown in
Figure 4.1. Starch hydrolysis using amylases would allow increasing the amount of L-
Chapter 4
78
lysine produced dramatically. In this case, a fed batch fermentation would most likely be
optimal, since C. glutamicum would react to an excess of carbon supply with a likely
increased byproduct formation as shown for the L-lysine producing strain C. glutamicum
ATCC 21513 (Hadj Sassi et al., 1998). Despite the significant difference in silage
composition the biomass yield was very similar to grass silage (Table 4.7) whereas L-
lysine yield was slightly lower.
Chapter 4
79
Fig. 4.2: Batch cultivation of lysine producing C. glutamicum SL on 1:4 diluted corn-
silage juice in the bioreactor using impeller 1 with 800 rpm. (A) Concentrations of
substrates maltose (Mal), glucose (Glu) and lactate (Lac), of product L-lysine (Lys)
0 2 4 6 8 10 12 140
20
40
60
80
100
120
0
2
DO
[%]
Time [h]
lnO
D66
0
Chapter 4
80
and of biomass provided as cell dry weight (CDW). (B) Oxygen uptake rate (OUR)
and carbon dioxide production rate (CPR). (C) Optical density (OD) and dissolved
oxygen concentration (DO – full line) given in % air saturation and stirrer and
oxygen adjustment. Dashed lines indicate identified growth phases I–III.
Shake flask cultivation using undiluted silage juice
To study the culture behavior with undiluted silage juice under oxygen limitation we
applied shake flask cultivations. These cultivations were performed using flasks with
fluorescence based dissolved oxygen sensing (Schneider et al., 2010). Both cultivations, on
grass silage juice and on corn silage juice show two different phases – one non-limiting
and one limiting concerning oxygen, as displayed in Figure S 4.2 and S 4.3. Non-limiting
growth in shake flasks was comparable to non-limited growth observed in bioreactors for
both silage juices. The same was observed for oxygen limiting conditions on both silage
juices. Our developed strain C. glutamicum SL, did not show any reduction in specific
growth rate or any other kind of lactic acid induced stress contrary to previous studies
(Seletzky et al., 2006). Product and biomass formation as well as L-lysine yield in shake
flasks were very similar on both types of silage (see Table 4.7). Even in comparison to
bioreactor cultivations applying oxygen sufficient and limiting conditions comparable
yields were obtained (Table 4.7). Compared to the precedent mutant strain C. glutamicum
lysCfbr dldPsod pycPsod malEPsod, with a selectivity of 0.067 g L-lysine (g biomass)-1, the L-
lysine selectivity of the new strain increased about four times.
Connections between growth, substrate consumption and L-lysine production are
summarized in concentration biomass plots of all cultivations shown in Figure 4.3. In all
cases the substrates fructose, glucose, lactate and maltose were consumed simultaneously.
The time point of depletion of the substrates was depending on their initial concentration.
The selectivity for L-lysine as indicated by the slope of the L-lysine curve was nearly
constant in both corn silage fermentations (Figure 4.3C and D) and in most parts of the
shake flask grass silage cultivation (Figure 4.3B), whereas in the grass silage fermentation
in the bioreactor depicted in Figure 4.1 it increased towards the end of cultivation (Figure
4.3A).
Chapter 4
81
Fig. 4.3: Relation of substrate and product concentrations and biomass production in
L-lysine producing C. glutamicum SL cultures on grass and corn silages. (A) Diluted
grass silage with raw data shown in Fig. 4.1. (B) Undiluted grass silage cultivated in
shake flasks (Figure S 4.2). (C) Diluted corn silage with raw data shown in Fig. 4.2.
(D) Undiluted corn silage cultivated in shake flasks (Figure S 4.3). Concentrations of
substrates fructose (Fru), maltose (Mal), glucose (Glu) and lactate (Lac), of product
L-lysine (Lys) and of biomass provided as cell dry weight (CDW). Total carbon (total
C) comprises only those substrates shown in the respective part of the figure.
Despite the differences of the silage juices regarding available carbon sources, the created
mutant strain C. glutamicum SL showed robust and flexible production of L-lysine. It was
Tota
l C [C
-m
M]
Lac
[C -
mM
]G
lu [C
-m
M]
Mal
[C -
mM
]
Lys
[C -
mM
]
Tota
l C [C
-m
M]
Fru
[C -
mM
]G
lu [C
-m
M]
Lac
[C -
mM
]
Lys
[C -
mM
]
Lys
[C -
mM
]
Tota
l C [C
-m
M]
Lac
[C -
mM
]G
lu [C
-m
M]
Fru
[C -
mM
]
Lys
[C -
mM
]
Tota
l C [C
-m
M]
Glu
c [ C
-mM
]La
c [ C
-mM
]M
al [
C-m
M]
Chapter 4
82
growing fast, with an almost constant L-lysine selectivity of 0.25 ± 0.03 g L-lysine (g
biomass)-1 and L-lysine carbon yields of slightly below 10% tolerating substantial changes
in the profile of available carbon sources, cultivation methods and oxygen supply. This is
very important, since the carbon source composition in silage juices is strongly varying,
depending on factors like the used green biomass, ensiling technique and conditions as
well as effects of wilting and drying (Hirst and Ramstad, 1957). With a secreted L-lysine
concentration of 4.9 g L-1 and additionally 0.5 g L-1 contained in the cellular biomass,
estimated from 17 g L-1 using data of Wittmann and de Graaf (2005), the demanded L-
lysine supplementation of 0.4% of total protein to optimize the nutritional value of silage
would be easily achievable. A comparison between the used silages with varying
composition and the performance of the strains under the various cultivation conditions is
shown in Table 4.7.
Chapter 4
83
Tab
le 4
.7:
Com
pari
son
of t
he u
sed
sila
ge j
uice
s an
d th
e pe
rfor
man
ce o
f th
e st
rain
s un
der
the
vari
ous
culti
vatio
n co
nditi
ons.
n 3 2 2 2
BR
– b
iore
acto
r, SF
– sh
ake
flask
s, µ
– sp
ecifi
c gr
owth
rate
, YP/
S – c
arbo
n yi
eld
of L
-lysi
ne o
n su
bstra
tes,
carb
on y
ield
of L
-lysi
ne o
n
subs
trate
s, Y
X/S
– c
arbo
n yi
eld
of b
iom
ass o
n su
bstra
tes.
n –
num
ber o
f exp
erim
ents
. Car
bon
yiel
ds a
re c
alcu
late
d co
nsid
erin
g al
l
subs
trate
s mea
sure
d, g
luco
se, f
ruct
ose,
mal
tose
, lac
tic a
cid
as d
epic
ted
in F
igur
e 4.
3 as
wel
l as q
uina
te, f
umar
ate,
glu
tam
ate,
ala
nine
,
leuc
ine
and
isol
euci
ne li
sted
in T
able
4.6
.
YX
/S
[C-m
mol
(C-m
ol)-1
]
380
± 8
340
± 6
385±
9
362
± 8
YP/
S
[C-m
mol
(C-m
ol)-1
]
94 ±
2
86 ±
3
87 ±
2
99 ±
3
µ [h
-1]
0.36
± 0
.01
0.35
± 0
.01
0.35
± 0
.01
0.35
± 0
.01
BR
SF
BR
SF
Mai
n C
arbo
n So
urce
s [g
L-1]
Lac
tic a
cid
40.0
± 7
.5
30.0
± 2
.0
Mal
tose
-
1.5
± 0.
5
Fruc
tose
20.0
± 4
.0
-
Glu
cose
4.0
± 2.
0
2.0
± 1.
5
Gra
ss
Cor
n
Chapter 4
84
This work indicates the robustness of the engineered organism C. glutamicum SL being
able to adapt to varying oxygen availabilities without major drawbacks concerning the L-
lysine producing ability, indicated by almost constant L-lysine yields under the varying
oxygen conditions investigated. This also holds for different silage juices obtained from
different sources which varied significantly in their composition.
Chapter 4
85
Concluding remarks
Sustainable economic growth requires safe, sustainable resources for industrial productions
as well as innovative ideas and technologies. Large scale ensiling of green biomass for the
production of methane opens up new substrate opportunities for biotechnological
production and assures a supply throughout the year, independent on sugar market price. In
this regard, the present work describes the engineering of C. glutamicum into efficient, fast
growing L-lysine producers using silage juice as a novel, renewable substrate. An already
created mutant (Neuner and Heinzle, 2011) seemed highly suitable for further engineering.
The performed pathway analysis using the network model and analysis method of Neuner
and Heinzle (Neuner and Heinzle, 2011) with inclusion of fructose as additional substrate
allowed to identify two additional targets, fructose 1,6-bisphosphatase and glyceraldehyde
3-phosphate. The first was already shown to increase L-lysine yield by increasing NADPH
availability and reducing byproduct formation. The overexpression of gapX caused a
reduction of formation of the byproduct dihydroxyacetone. The resulting strain,
C.glutamicum SL was stable and fast growing. It clearly showed an improved performance
compared to the precedent strain exhibiting a threefold increased L-lysine yield and a
drastic reduction of dihydroxyacetone formation. These characteristics make the C.
glutamicum lysCfbr dldPsod pycPsod malEPsod fbpPsod gapXPsod mutant a very versatile, adaptive
strain and show the potential of silage as a renewable, sustainable substrate for
biotechnological applications. The obtained carbon yields of 86 C-mmol (C-mol)-1 to 94 C-
mmol (C-mol)-1 are below those reported in the literature for high producing strains but
provide a good starting point for extended engineering using further well known genetic
modifications (Becker and Wittmann, 2011; Becker and Wittmann, 2012; Becker et al.,
2011). Future work regarding L-lysine production on silage is directed towards further
increase of L-lysine yield, selectivity and product titer. L-lysine production was hardly
effected by low oxygen supply. This is very interesting for an efficient production creating
reduced aeration cost both in terms of investment as well as operation costs. Summing up,
this work shows that silages are promising raw materials for biotechnological production
using C. glutamicum.
Chapter 4
86
Acknowledgements. We thank the Fachagentur für Nachwachsende Rohstoffe (FNR) who
supported this work (FKZ 22004908). KS acknowledges support by BMBF (Federal
Ministry of Education and Research - Germany, Project SWEEPRO, FKZ 0315800B). We
are very grateful for the most valuable support by Michel Fritz concerning instrumental
analysis.
Chapter 5
87
Chapter 5
Pretreatment of corn silage juice
Chapter 5
88
Corn silage contains considerable amounts of starch (~300 g L-1), the major storage
polysaccharide of cereal grains. Starch is a substrate not metabolizable by C. glutamicum
wilde type due to the absence of starch degrading enzymes (Seibold et al., 2006; Tateno et
al., 2007). In order to assess the possibilities for starch utilization, we investigated special
pretreatment methods for corn silage juice. A schematic overview is depicted in Figure 5.1.
Fig. 5.1: Schematic overview of the corn silage pretreatment. Pretreatment steps are
highlighted in grey boxes.
Enzymatic starch hydrolysation of the pressed silage juice using α-amylase from
Aspergillus oryzae (Sigma, Switzerland) and amyloglucosidase from Aspergillus niger
(Sigma, Switzerland) reduced the size of the starch grains and degraded the available
Chapter 5
89
starch to maltose and glucose. During the procedure, glucose and maltose concentrations
were determined using HPLC. The morphology of the starch grains was documented using
an IX70 microscope (Olympus, Hamburg, Germany).The results are shown in Table 5.1.
Tab
le 5
.1: P
retr
eatm
ent o
f cor
n si
lage
juic
e. S
tarc
h hy
drol
ysis
usi
ng α
-am
ylas
e an
d am
ylog
luco
sida
se.
A
myl
oglu
cosi
dase
-Tre
atm
ent
10.9
151.
1
α-A
myl
ase-
Tre
atm
ent
40.1
8.7
No
Pret
reat
men
t
0 0
Mor
phol
ogy
Con
cent
ratio
n [g
L-1
]
Mal
tose
Glu
cose
Chapter 5
90
During the initial phase, starch grains are clearly visible and no glucose or maltose is
detectable. After the use of α-amylase for 40 h, the size of the starch grains is significantly
altered, the grains are smaller and a maltose concentration of 40.1 g L-1 was detected, with
a glucose concentration of 8.7 g L-1. The analysis of the sample after the use of
amyloglucosidase for another 40 h clearly shows the influence of the enzyme regarding
starch grain morphology and the concentrations of glucose and maltose. The grains are
barely visible, with a maltose concentration of 10.9 g L-1 and 151 g L-1 glucose in the
supernatant. Since the initial starch concentration in the used corn silage juice was 214 g L-
1, the enzymatic treatment converted 75% of the starch into maltose and glucose, making
the carbon available for C. glutamicum. This is very beneficial for two different reasons.
First, there is a lot more carbon available for L-lysine fermentations. Second, when starch
supplied from corn silage is fed to animals, incomplete digestion of non-fiber
polysaccharides like starch appears to be very common for lactating dairy cows and other
ruminants (Eastridge et al., 2011). Rapid fermentation of starch in the rumen increases the
acidity in this compartment and the risk of digestive disorders, particularly ruminal
acidosis (Johnson et al., 2003). An enzymatic pretreatment would convert a high
percentage of the carbon from starch to substrates available for C. glutamicum that can be
converted to L-lysine. At the same time, the final starch concentration in the fodder is
reduced and the incomplete digestion, where around 25% of the energy contained in starch
would be lost (Huntington, 1997), can be avoided. This would significantly improve the
economic efficiency of the process. For all the cultivation experiments performed with
corn silage juice, the starch fraction has not been used.
Chapter 6
91
Chapter 6
Concluding discussion and remarks
Chapter 6
92
Concluding discussion
The depletion of fossil resources together with a constantly growing population and
environmental issues like global warming and increasing waste problems call for
sustainable alternatives regarding energy, fuels and chemical production. Solar, wind and
water energy are already exploited as alternatives, but can only partially replace the fossil
energy. For the production of transportation fuels and chemicals, the utilization of plant
biomass shows the greatest potential. Until now, renewable raw materials had an economic
disadvantage, especially if it comes to the production of bulk products, as they are priced
higher than fossil resources, compared to their respective product yield, sophisticated
logistics due to low transport densities and short shelf life. These facts impeded the
development of bio-based processes, leaving fossil raw materials the undisputed primacy
as an industrial feedstock. However, expected price increases of crude oil and the limited
availability of these resources are profoundly challenging this primacy. As a result of this
development, new technologies based on plant biomass, aiming at multi-product plants
with increased efficiency of biomass utilization arised. Using low priced materials and
utilizing these resources to the utmost possible degree will eventually compensate for the
initial economic disadvantage of bio-based feedstocks, allowing these technologies to
compete with conventional processes based on fossil resources. It is obvious that technical
and economic barriers must be conquered for an efficient utilization of these feedstocks.
Many of the economic problems will be solved, when the technical issues are resolved.
Additionally, the price increase of petroleum based resources will also contribute to the
economic feasibility of bio-based products.
In this thesis, the potential use of silage and silage juice for the biotechnological
production of the amino acid L-lysine was investigated. Three different aspects regarding
amino acid production on silage have been addressed. First, in order to make a bio-based
process efficient, the substrate spectrum, anabolic power supply in form of NADPH and
precursor availability were optimized. By performing a metabolic network analysis,
specific targets for genetic engineering were identified. Second, the treatment of silage
juice for optimal use and the influence of process parameters like aeration and stirring rate
on production efficiency were elucidated. Finally, L-lysine production on different silage
juices was adressed.
Chapter 6
93
Expansion of the substrate spectrum and L-lysine production on
synthetic silage juice
Two major compounds of silage juice are lactate and glucose. It was of interest to
investigate the use of such a type of raw material, consisting of considerable amounts of
lactate and glucose, for the biotechnological production of L-lysine. Growth of C.
glutamicum on mixtures of glucose and racemic lactate has not been studied in detail
before and attempts to overproduce L-lysine have not been reported yet using such
substrates. C. glutamicum wild type shows poor growth on D-lactate and racemic lactate
mixtures (Scheer et al., 1988). As thoroughly described in Chapter 3, overexpression of
the constitutively expressed D-lactate dehydrogenase (dld) expanded the substrate
spectrum and improved growth. A production strain capable of completely metabolizing
carbon sources contained in silage juice would be desirable, in order to increase the carbon
utilization of a production host. The engineered strain is currently able to utilize D-lactate
in addition to its usual carbon sources, resulting in an increased carbon efficiency and fast
growth. The overexpression of dld significantly improved growth on racemic lactate with
respect to growth rate (55% increase compared to parental strain) as well as biomass yield
(17% increase). The increase in growth rate is more than could be expected from a simple
duplication of lactate supply by dld overexpression and represents a good starting point for
further metabolic engineering. Although this strain shows no detectable, growth coupled L-
lysine production, it can be further optimized regarding NADPH supply and precursor
availability for L-lysine production. By implementing a network analysis, the maximum
theoretical yield of L-lysine on glucose and lactate mixtures was calculated based on the
stoichiometry of the metabolic network of C. glutamicum. The highest L-lysine carbon
yield of an elementary flux mode with simultaneous consumption of glucose and lactate
was 84.4%. However, this is a so called extreme mode, producing exclusively L-lysine,
with no biomass formation. Regarding modes with simultaneous glucose and lactate
consumption as well as simultaneous biomass and L-lysine biosynthesis, the maximum
theoretical carbon yield is 53% with an associated biomass carbon yield of 19%. The
elementary flux mode analysis combined with useful objective functions (see Chapter 3)
permitted a valuable prediction of potential targets for increasing L-lysine production by
overexpression of selected genes. This analysis predicted stimulation of L-lysine
Chapter 6
94
production by a combined overexpression of pyruvate carboxylase and malic enzyme as
the most promising targets, followed by gluconeogenic enzymes and enzymes of the
pentose phosphate pathway. Compared to the parental strain, C. glutamicum lysCfbr,
exhibiting a growth rate of 0.13 h-1 on glucose-lactate mixtures, the sole overexpression of
D-lactate dehydrogenase (dld) in C. glutamicum lysCfbrdldPsod increased the growth rate by
more than 100% (µ = 0.28 h-1). The additional overexpression of pyruvate carboxylase
(pyc) further increased the growth rate up to 0.39 h-1.One possible reason why the parental
strain C. glutamicum lysCfbr did not produce detectable amounts of L-lysine neither on
lactate, nor on glucose-lactate mixtures, seems to be the availability of NADPH. A fact
supporting this interpretation is that lactate uptake increased after the overexpression of D-
lactate dehydrogenase (dld), but still no L-lysine was produced. Overexpression of
pyruvate carboxylase (pyc) and malic enzyme (malE) enabled L-lysine production on
synthetic silage juice, by increasing the precursor supply in form of oxaloacetate and
providing enough NADPH for biomass and L-lysine production. The resulting strain C.
glutamicum lysCfbrdldPsodpycPsodmalEPsod, with a selectivity of 0.067 g L-lysine (g
biomass)−1 when cultivated on silage juice is a promising target for further engineering.
The positive effects facilitated by the modified gene expression regarding growth and L-
lysine production on synthetic silage juice prove the usefulness of combining bioinformatic
methods with wet lab experiments.
L-lysine production on silage juices
In Chapter 4, the application of silage juice for biotechnological use has been
investigated. Sterilization of the silage juice and the avoidance of inhibitory compounds
due to the occurring Maillard reaction are a crucial step for the use of silage juice as a
fermentation substrate. Different pretreatment methods for grass and corn silage have been
discussed, according to the respective silage juice characteristics. Regarding corn silage
juice, the high amount of starch (~ 300 g L-1) is not available for C. glutamicum, since it is
not able to directly metabolize starch. The starch represents a substantial amount of carbon
and it is of major interest to investigate ways to channel it into L-lysine production, as
described in Chapter 5. With a maltose concentration of 10.9 g L-1 and 151 g L-1 glucose
Chapter 6
95
in the supernatant after enzymatic treatment, valuable carbon sources became readily
available. These findings open up new possibilities for biotechnological use. The high
glucose concentration would be suitable for fed batch cultivation, providing a good
perspective for L-lysine production. The downside is the more sophisticated fermentation
equipment and the impossibility of a simple fermentation that can be performed directly on
farming units as well as the economic factor. The additional use of enzymes is a cost factor
that must be taken into account. A different option would be co-fermentation using
microorganisms secreting starch degrading enzymes or even the use of engineered C.
glutamicum strains, displaying α-amylase from e.g Streptococcus bovis on its cell surface
(Tateno et al., 2007). Heterologuous gene expression would imply that the constructed
strain belongs to the GMO category. This means that the use of L-lysine as a feed
supplement would no longer be an option. In this context, heterologuous expression of
genes involved in pentose metabolism (xylose) can also be mentioned. Silage contains
xylose, in amounts between 5-8 g L-1, depending on the used biomass. Xylose cannot be
metabolized by C. glutamicum wild type (Kawaguchi et al., 2006). This genetic
modification would increase the amount of available carbon, at the same time excluding
the optional use of L-lysine as a feed supplement under the prevailing conditions for
acceptance of GMO’s. Besides the investigation of proper methods for silage juice
treatment, the strains described in Chapter 3 have been further optimized towards L-lysine
production, improved growth as well as reduced byproduct formation and tested on grass
and corn silage juices. Two additional targets for overexpression (fbp and gapX) that were
predicted by the same model and method used in Chapter 3, but using lactate, glucose and
fructose as substrates, were investigated. Overexpression of the fructose 1,6-
bisphosphatase significantly improved the L-lysine yield on grass and corn silage juice,
while the byproduct formation was reduced. Additional overexpression of glyceraldehyde
3-phosphate dehydrogenase further diminished the byproduct formation. Keeping in mind
that one of the top challenges regarding bio-based products is the economic production,
different cultivation setups have been tested, in order to assess the influence and necessity
of aeration and oxygen supply during the fermentation. All these factors contribute to the
estimation of capital and operating costs, determining the feasibility of the process
(Heinzle et al., 2006). C. glutamicum SL performed well under the tested cultivation
setups, not showing any reduction in specific growth rate or any other kind of lactic acid
induced stress due to high concentrations in undiluted silage juices. Product and biomass
formation as well as L-lysine yield in shake flasks were very similar on both types of
Chapter 6
96
silage. Even in comparison to bioreactor cultivations applying oxygen sufficient and
limiting conditions comparable yields were obtained, showing the robustness of the created
strain. It can adapt to varying oxygen availabilities without major drawbacks concerning
the L-lysine producing ability, indicated by almost constant L-lysine yields under the
varying oxygen conditions investigated. A detailed description and discussion of these
findings is available in Chapter 4. This also holds for the productivity on different silage
juices which varied significantly in their carbon source composition. The obtained L-lysine
carbon yields of 86–94 C-mmol (C-mol)−1 and a L-lysine selectivity of 0.25 ± 0.03 g L-
lysine (g biomass)−1 at growth rates of 0.35 ± 0.01 h-1 are a promising starting point for
extended engineering. All performed genetic modifications are genome based, without the
need of any selectivity markers like antibiotic resistance remaining in the strains. This
makes the C. glutamicum SL strain suitable and very interesting candidate for the feed and
farming industry, where L-lysine supplementation is of major interest. This fact and the
GRAS classification of C. glutamicum allow the use of the complete fermentation broth,
containing the microorganisms, as a feed supplement. The mentioned fodder
supplementation with 0.4% L-lysine (as described in Chapter 4) in order to increase the
PER of corn protein is already achieved. This is a considerable advantage compared to
conventional processes for crystalline L-lysine*HCl production where the accumulated
biomass, byproducts and residues from the fermentation process must be disposed.
Purification of the L-lysine using ion exchange chromatography is also associated with a
loss of considerable amounts of the product. Conventional amino acid production is a
costly process, requiring the handling of hazardous substances with a high ecological
burden like ammonia solution as an eluent for the ion exchange columns and hydrochloric
acid for the neutralization of the L-lysine base (Kelle et al., 2005). The idea presented here
describes a simpler route to feed supplementation which avoids the ecological and
technical burden of existing methods. Besides that, L-lysine is available as a natural, free
amino acid with a high biological efficacy, produced exclusively from renewable resources
without any additive.
Chapter 6
97
Chapter 6
98
Concluding remarks
This thesis investigates the use of silage and silage juice as renewable fermentation
substrates for L-lysine production with C. glutamicum. In particular, we showed the great
potential of L-lysine overproduction as predicted by the performed metabolic network
analysis. The determination of the maximal theoretical yields is very promising, since it is
the evaluation of the feasibility of such a process. The performed elementary mode
analysis was very useful for the identification of potential targets for genetic engineering,
as demonstrated by the implementation of the findings in Chapters 3 and 4. A widened
substrate spectrum and the associated improved carbon efficiency and growth
characteristics were obtained. It became obvious, that a crucial step in L-lysine
overproduction, especially on substrates employing gluconeogenic pathways, is the supply
of anabolic reduction power in form of NADPH. The improved NADPH availability due to
the overexpression of malic enzyme was proven to be of utmost importance, directly
resulting in improved L-lysine yields. The constructed strains were further engineered
towards better L-lysine selectivity, with increased L-lysine yields and a lower byproduct
formation. A very important part of the thesis was a substrate oriented approach, where the
chosen substrates, grass and corn silage juice, were tested for optimal use as an additive
free, natural fermentation substrates. A new and innovative process for the
biotechnological production of L-lysine was tested, where the whole manufacturing
concept including the production strain, used raw materials and cultivation setups were
systematically tailored for optimal environmental compatibility, reduced handling of
hazardous materials as well as minimal resource depletion and waste generation. With a
reduced investment outlay and being easily applicable on agricultural and farming units,
the L-lysine fermentation on silage juices is highly attractive, in ecological and economic
terms. By increasing the nutritional value of fodder, L-lysine fermentations on silage juices
are a very promising, cost effective contribution to a sustainable economy, more
environmentally compatible by the reduced nitrogen emission levels.
References
99
References Amon, T., Amon, B., Kryvoruchko, V., Zollitsch, W., Mayer, K. and Gruber, L. (2007).
Biogasproduction from maize and dairy cattle manure:Influence of biomass composition on the methane yield. Agric, Ecosyst Environ 118, 173‐182.
Angelidaki, I., Alves, M., Bolzonella, D., Borzacconi, L., Campos, J. L., Guwy, A. J., Kalyuzhnyi, S.,
Jenicek, P. and van Lier, J. B. (2009). Defining the biomethane potential (BMP) of solid organic wastes and energy crops: a proposed protocol for batch assays. Water Sci Technol 59, 927‐934.
Bartlett, A. T. M. and White, P. J. (1985). Species of Bacillus that make a vegetative peptidoglycan
containing lysine lack diaminopimelate epimerase but have diaminopimelate dehydrogenase J Gen Microbiol 131, 2145‐2152.
Becker, J., Klopprogge, C., Schröder, H. and Wittmann, C. (2009). TCA cycle engineering for
improved lysine production in Corynebacterium glutamicum. Appl Environ Microbiol 75, 7866‐7869.
Becker, J., Klopprogge, C., Zelder, O., Heinzle, E. and Wittmann, C. (2005). Amplified expression of
fructose 1,6‐bisphosphatase in Corynebacterium glutamicum increases in vivo flux through the pentose phosphate pathway and lysine production on different carbon sources. Appl Environ Microbiol 71, 8587‐8596.
Becker, J. and Wittmann, C. (2011). Bio‐based production of chemicals, materials and fuels ‐
Corynebacterium glutamicum as versatile cell factory. Curr Opin Biotechnol 23, 631‐640. Becker, J. and Wittmann, C. (2012). Systems and synthetic metabolic engineering for amino acid
production ‐ the heartbeat of industrial strain development. . Curr Opin Biotechnol. 23, 718‐726.
Becker, J., Zelder, O., Hafner, S., Schröder, H. and Wittmann, C. (2011). From zero to hero‐design‐
based systems metabolic engineering of Corynebacterium glutamicum for L‐lysine production. Metab Eng 13, 159‐168.
Belitz, H.‐D., Grosch, W., Schieberle, P. (2001). Lehrbuch der Lebensmittelchemie. Springer Verlag:
Berlin. Benson, D. A., Karsch‐Mizrachi, I., Lipman, D. J., Ostell, J. and Wheeler, D. L. (2008). GenBank.
Nucleic Acids Res 36, D25‐D30. Bentley, R. W. (2002). Global oil & gas depletion:an overview. Energy Policy 30, 189–205. Boehmel, C., Lewandowski, I. and Claupein, W. (2008). Comparing annual and perennial energy
cropping systems with different management intensities. Agricultural Systems 96, 224–236.
Bonamy, C., Labarre, J., Cazaubon, L., Jacob, C., Le Bohec, F., Reyes, O. and Leblon, G. (2003). The
mobile element IS1207 of Brevibacterium lactofermentum ATCC21086: isolation and use in the construction of Tn5531, a versatile transposon for insertional mutagenesis of Corynebacterium glutamicum. J Biotechnol 104, 301‐309.
References
100
Bott, M. and Niebisch, A. (2003). The respiratory chain of Corynebacterium glutamicum. J Biotechnol 104, 129‐153.
Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein‐dye binding. Anal Biochem 72, 248‐254.
Burkovski, A. (2008). Corynebacteria ‐ Genomics and molecular biology. Caister Academic Press:
Norfolk, UK. Carlson, R. and Srienc, F. (2004a). Fundamental Escherichia coli biochemical pathways for biomass
and energy production: creation of overall flux states. Biotechnol Bioeng 86, 149‐162. Carlson, R. and Srienc, F. (2004b). Fundamental Escherichia coli biochemical pathways for biomass
and energy production: identification of reactions. Biotechnol Bioeng 85, 1‐19. Chen, Z., Meyer, W., Rappert, S., Sun, J. and Zeng, A. P. (2011). Coevolutionary analysis enabled
rational deregulation of allosteric enzyme inhibition in Corynebacterium glutamicum for lysine production. Appl Environ Microbiol 77, 4352‐4360.
Clark, J. H. (2007). Green chemistry for the second generation biorefinery ‐ sustainable chemical
manufacturingbased on biomass. J Chem Technol Biotechnol 82, 603‐609. Cocaign‐Bousquet, M. and Lindley, N. (1995). Pyruvate overflow and carbon flux withinthe central
metabolic pathways of C.glutamicum during growth on lactate. Enzyme Microb Technol 17, 260‐267.
Cocaign, M., Monnet, C. and Lindley, N. D. (1993). Batch kinetics of Corynebacterium glutamicum
during growth on various carbon substrates: use of substrate mixtures to localise metabolic bottlenecks. Appl Microbiol Biotechnol 40, 526‐530.
Cremer, J., Eggeling, L. and Sahm, H. (1991). Control of the lysine biosynthetic sequence in
Corynebacterium glutamicum as analyzed by overexpression of the individual corresponding genes. Appl Biochem Biotechnol 57, 1746‐1752.
Crow, V. L. and Wittenberger, C. L. (1979). Separation and properties of NAD+‐ and NADP+‐
dependent glyceraldehyde‐3‐phosphate dehydrogenases from Streptococcus mutans. J Biol Chem 254, 1134‐1142.
Dale, B. E. (2003). ‘Greening’ the chemical industry: research and development priorities for
biobased industrial products. J Chem Technol Biotechnol 78, 1093–1103. Datta, R., Tsai, S.‐P., Bonsignore, P., Moon, S.‐H. and Frank, J., R. (1995). Technological and
economic potential of poly lactic acid and lactic acid derivates. FEMS Microbiol Rev 16, 221‐231.
Demain, A. (2000). Microbial biotechnology. Trends Biotechnol 18, 26‐31. Eastridge, M. L., Lefeld, A. H., Eilenfeld, A. M., Gott, P. N., Bowen, W. S. and Firkins, J. L. (2011).
Corn grain and liquid feed as nonfiber carbohydrate sources in diets for lactating dairy cows. J Dairy Sci 94, 3045‐3053.
References
101
Eggeling, L. and Bott, M. (2005). Handbook of Corynebacterium glutamicum. Taylor & Francis Group: Boca Raton.
Eikmanns, B. J. (2005). Central Metabolism: tricarboxylic acid cycle and anaplerotic reactions in
Eggeling, L. and Bott, M. (eds), Handbook of Corynebacterium glutamicum, Handbook of Corynebacterium glutamicum, CRC Press: Boca Raton, pp. 241‐276.
Ensari, S. and Lim, H. C. (2003). Kinetics of L‐lysine fermentation: a continuous culture model
incorporating oxygen uptake rate. Appl Microbiol Biotechnol 62, 35‐40. Ericsson, K. and Nilsson, L. J. (2006). Assessment of the potential biomass supply in Europe using a
resource‐focused approach. Biomass and Bioenergy 30, 1‐15. Gavrilescu, M. and Chisti, Y. (2005). Biotechnology—a sustainable alternative for chemical
industry. Biotechnology Advances 23, 471–499. Georgi, T. (2006). Regulation der Zuckerverwertung in C. glutamicum, Berichte des
Forschungszentrums Jülich. Georgi, T., Engels, V. and Wendisch, V. F. (2008). Regulation of L‐lactate utilization by the FadR‐
type regulator LldR of Corynebacterium glutamicum. J Bacteriol 190, 963‐971. Georgi, T., Rittmann, D. and Wendisch, V. F. (2005). Lysine and glutamate production by
Corynebacterium glutamicum on glucose, fructose and sucrose: roles of malic enzyme and fructose‐1,6‐bisphosphatase. Metab Eng 7, 291‐301.
Gourdon, P., Baucher, M. F., Lindley, N. D. and Guyonvarch, A. (2000). Cloning of the malic
enzyme gene from Corynebacterium glutamicum and role of the enzyme in lactate metabolism. Appl Environ Microbiol 66, 2981‐2987.
Hadj Sassi, A., Fauvart, L., Deschamps, A., M. and Lebeault, M. (1998). Fed‐batch production of l‐
lysine by Corynebacterium glutamicum. Biochem Eng J 1, 85‐90. Hall, D. O. and Scrase, J. I. (1998). Will biomass energy be the environmentally friendly fuel of the
future? Biomass and Bioenergy 15, 357–367. Hanahan, D. (1983). Studies on transformation of Escherichia coli with plasmids. J Mol Biol 166,
557‐80. Hanegraaf, M. C., Biewinga, E. and van der Bijl, G. (1998). Assessing the ecological and economic
sustainability of energy crops. Biomass and Bioenergy 15, 345‐355. Heinzle, E., Biwer, A. and Cooney, C. (Eds) (2006). Development of sustainable bioprocesses:
modeling and assessment, John Wiley and Sons: West Sussex, England. Heinzle, E., Oeggerli, A. and Dettwiler, B. (1990). On‐line fermentation gas analysis: error analysis
and application of mass spectrometry. Analytica Chimica Acta 238, 101‐115. Heinzle, E., Yuan, Y., Kumar, S., Wittmann, C., Gehre, M., Richnow, H. H., Wehrung, P., Adam, P.
and Albrecht, P. (2008). Analysis of 13C labeling enrichment in microbial culture applying metabolic tracer experiments using gas chromatography‐combustion‐isotope ratio mass spectrometry. Anal Biochem 380, 202‐210.
References
102
Hermann, B. G. and Patel, M. (2007). Today's and tomorrow's bio‐based bulk chemicals from white biotechnology: a techno‐economic analysis. Appl Biochem Biotechnol 136, 361‐388.
Hill, K. (2000). Fats and oils as oleochemical raw materials. Pure Appl Chem 72, 1255–1264. Hirst, E. L. and Ramstad, L. (1957). Changes in organic acid content of perennial rey grass during
conservation. J Sci Food Agric, 727‐732. Hoffert, M. I., Caldeira, K., Jain, A. K., Haites, E. F., Harvey, L. D. D., Potter, S. D., Schlesinger, M. E.,
Schneider, S. H., Watts, R. G., Wigley, T. M. L. and Wuebbles, D. J. (1998). Energy implications of future stabilization of atmospheric CO2 content. Nature 395, 881‐884.
Huntington, G. B. (1997). Starch utilization by ruminants: from basics to the bunk. J Anim Sci 75,
852‐867. Ikeda, M. (2003). Amino acid production processes. Adv Biochem Eng Biotechnol 79, 1‐35. Ikeda, M. and Katsumata, R. (1992). Metabolic engineering to produce tyrosine or phenylalanine
in a tryptophan‐producing Corynebacterium glutamicum strain. Appl Environ Microbiol 58, 781‐785.
Ikeda, M. and Nakagawa, S. (2003). The Corynebacterium glutamicum genome: features and
impacts on biotechnological processes. Appl Microbiol Biotechnol 62, 99‐109. Ikeda, M., Ohnishi, J., Hayashi, M. and Mitsuhashi, S. (2006). A genome‐based approach to create
a minimally mutated Corynebacterium glutamicum strain for efficient L: ‐lysine production. J Ind Microbiol Biotechnol 33, 610‐5.
Inoue, H., Nojima, H. and Okayama, H. (1990). High efficiency transformation of Escherichia coli
with plasmids. Gene 96, 23‐28. Inui, M., Kawaguchi, H., Murakami, S., Vertes, A. A. and Yukawa, H. (2004a). Metabolic
engineering of Corynebacterium glutamicum for fuel ethanol production under oxygen‐deprivation conditions. J Mol Microbiol Biotechnol 8, 243‐254.
Inui, M., Murakami, S., Okino, S., Kawaguchi, H., Vertes, A. A. and Yukawa, H. (2004b). Metabolic
analysis of Corynebacterium glutamicum during lactate and succinate productions under oxygen deprivation conditions. J Mol Microbiol Biotechnol 7, 182‐96.
Jäger, W., Schäfer, A., Pühler, A., Labes, G. and Wohlleben, W. (1992). Expression of the Bacillus
subtilis sacB gene leads to sucrose sensitivity in the gram‐positive bacterium Corynebacterium glutamicum but not in Streptomyces lividans. J Bacteriol 174, 5462‐5465.
Johnson, L. M., Harrison, J. H., Davidson, D., Hunt, C., Mahanna, W. C. and Shinners, K. (2003).
Corn silage management: effects of hybrid, maturity, chop length, and mechanical processing on rate and extent of digestion. J Dairy Sci 86, 3271‐3299.
Johnson, V. A. and Lay, C. (1974). Genetic improvement of plant protein. J Agr Food Chem 22, 558‐
566.
References
103
Kalinowski, J., Bathe, B., Bartels, D., Bischoff, N., Bott, M., Burkovski, A., Dusch, N., Eggeling, L., Eikmanns, B. J., Gaigalat, L., Goesmann, A., Hartmann, M., Huthmacher, K., Kramer, R., Linke, B., McHardy, A. C., Meyer, F., Mockel, B., Pfefferle, W., Puhler, A., Rey, D. A., Ruckert, C., Rupp, O., Sahm, H., Wendisch, V. F., Wiegrabe, I. and Tauch, A. (2003). The complete Corynebacterium glutamicum ATCC 13032 genome sequence and its impact on the production of L‐aspartate‐derived amino acids and vitamins. J Biotechnol 104, 5‐25.
Kamm, B. and Kamm, M. (2004). Principles of biorefineries. Appl Microbiol Biotechnol 64, 137‐145. Kamm, B. and Kamm, M. (2007). Biorefineries‐multi product processes. Adv Biochem Eng
Biotechnol 105, 175‐204. Kanehisa, M., Araki, M., Goto, S., Hattori, M., Hirakawa, M., Itoh, M., Katayama, T., Kawashima, S.,
Okuda, S., Tokimatsu, T. and Yamanishi, Y. (2008). KEGG for linking genomes to life and the environment. Nucleic Acids Res 36, D480‐D484.
Karp, P. D., Keseler, I. M., Shearer, A., Latendresse, M., Krummenacker, M., Paley, S. M., Paulsen,
I., Collado‐Vides, J., Gama‐Castro, S., Peralta‐Gil, M., Santos‐Zavaleta, A., Penaloza‐Spinola, M. I., Bonavides‐Martinez, C. and Ingraham, J. (2007). Multidimensional annotation of the Escherichia coli K‐12 genome. Nucleic Acids Res 35, 7577‐7590.
Kawaguchi, H., Vertes, A. A., Okino, S., Inui, M. and Yukawa, H. (2006). Engineering of a xylose
metabolic pathway in Corynebacterium glutamicum. Appl Environ Microbiol 72, 3418‐3428.
Kelle, R., Hermenn, T. and Bathe, B. (2005). L‐Lysine production In Eggeling, L. and Bott, M. (Eds),
Handbook of Corynebacterium glutamicum, Taylor&Francis Group: Boca Raton, pp. 465‐489.
Kiefer, P., Heinzle, E., Zelder, O. and Wittmann, C. (2004). Comparative metabolic flux analysis of
lysine‐producing Corynebacterium glutamicum cultured on glucose or fructose. Appl Environ Microbiol 70, 229‐239.
Kim, H. M., Heinzle, E. and Wittmann, C. (2006). Deregulation of aspartokinase by single
nucleotide exchange leads to global flux rearrangement in the central metabolism of Corynebacterium glutamicum. J Microbiol Biotechnol 16, 1174‐1179
Kind, S., Jeong, W. K., Schroder, H. and Wittmann, C. (2010). Systems‐wide metabolic pathway
engineering in Corynebacterium glutamicum for bio‐based production of diaminopentane. Metab Eng 12, 341‐351.
Kind, S., Kreye, S. and Wittmann, C. (2011). Metabolic engineering of cellular transport for
overproduction of the platform chemical 1,5‐diaminopentane in Corynebacterium glutamicum. Metab Eng 13, 617‐627.
Kinoshita, S., Udaka, S. and Shimono, M. (2004). Studies on the amino acid fermentation. Part 1.
Production of L‐glutamic acid by various microorganisms. J Gen Appl Microbiol 50, 331‐343.
Kircher, M. and Pfefferle, W. (2001). The fermentative production of L‐lysine as an animal feed
additive. Chemosphere 43, 27‐31.
References
104
Kirchner, O. and Tauch, A. (2003). Tools for genetic engineering in the amino acid‐producing bacterium Corynebacterium glutamicum. J Biotechnol 104, 287‐299.
Kiss, R. D. and Stephanopoulos, G. (1992). Metabolic characterization of a L‐lysine‐producing
strain by continuous culture. Biotechnol Bioeng 39, 565‐574. Kjeldsen, K. R. and Nielsen, J. (2009). In silico genome‐scale reconstruction and validation of the
Corynebacterium glutamicum metabolic network. Biotechnol Bioeng 102, 583‐597. Klapa, M. I., Aon, J. C. and Stephanopoulos, G. (2003). Systematic quantification of complex
metabolic flux networks using stable isotopes and mass spectrometry. Eur J Biochem 270, 3525‐3542.
Kohl, T. A. and Tauch, A. (2009). The GlxR regulon of the amino acid producer Corynebacterium
glutamicum: Detection of the corynebacterial core regulon and integration into the transcriptional regulatory network model. J Biotechnol 143, 239‐246.
Krajcovicova‐Kudlackova, M., Babinska, K. and Valachovichova, M. (2005). Health benefits and
risks of plant proteins. Bratisl Lek Listy 106, 231‐234. Krömer, J. O., Fritz, M., Heinzle, E. and Wittmann, C. (2005). In vivo quantification of intracellular
amino acids and intermediates of the methionine pathway in Corynebacterium glutamicum. Anal Biochem 340, 171‐173.
Krömer, J. O., Sorgenfrei, O., Klopprogge, K., Heinzle, E. and Wittmann, C. (2004). In‐depth
profiling of lysine‐producing Corynebacterium glutamicum by combined analysis of the transcriptome, metabolome, and fluxome. J Bacteriol 186, 1769‐1784.
Krömer, J. O., Wittmann, C., Schröder, H. and Heinzle, E. (2006). Metabolic pathway analysis for
rational design of L‐methionine production by Escherichia coli and Corynebacterium glutamicum. Metab Eng 8, 353‐369.
Kromus, S., Wachter, B., Koschuh, W., Mandl, M., Krotscheck, C. and Nardoslawsky, M. (2004).
The green biorefinery Austria ‐ Developement of an integrated system for green mass utilization. Chem Biochem Eng 18, 7 ‐ 12.
Krotscheck, C., Kromus, S., Novalin, S., Koschuh, W. and Hong, T. (2004). Grüne Bioraffinerie:
Gewinnung von Milchsäure aus Grassilagesaft. Bundesministerium für Verkehr, Innovation und Technologie: Wien.
Leclercq, B. (1998). Lysine: Specific effects of lysine on broiler production: comparison with
threonine and valine. Poultry Science 77, 118‐123. Leuchtenberger, W., Huthmacher, K. and Drauz, K. (2005). Biotechnological production of amino
acids and derivatives: current status and prospects. Appl Microbiol Biotechnol 69, 1‐8. Lichtenthaler, W. F. and Peters, S. (2004). Carbohydrates as green raw materials for the chemical
industry. Comptes Rendus Chimie 7, 65‐90. Liebl, W., Bayerl, A., Schein, B., Stillner, U. and Schleifer, K. H. (1989). High efficiency
electroporation of intact Corynebacterium glutamicum cells. FEMS Microbiol Lett 53, 299‐303.
References
105
Liebl, W., Ehrmann, M., Ludwig, W. and Schleifer, K. H. (1991). Transfer of Brevibacterium divaricatum DSM 20297T, "Brevibacterium flavum" DSM 20411, "Brevibacterium lactofermentum" DSM 20412 and DSM 1412, and Corynebacterium glutamicum and their distinction by rRNA gene restriction patterns. Int J Syst Bacteriol 41, 255‐260.
Liebl, W., Sinskey, A. J. and Schleifer, K. H. (1992). Expression, secretion, and processing of
staphylococcal nuclease by Corynebacterium glutamicum. J Bacteriol 174, 1854‐1861. Liu, Q., Ouyang, S. P., Kim, J. and Chen, G. Q. (2007). The impact of PHB accumulation on L‐
glutamate production by recombinant Corynebacterium glutamicum. J Biotechnol 132, 273‐279.
Lorenz, P., Liebeton, K., Niehaus, F. and Eck, J. (2002). Screening for novel enzymes for biocatalytic
processes: accessing the metagenome as a resource of novel functional sequence space. Curr Opin Biotechnol 13, 572‐577.
Madigan, M. T. and Marrs, B. L. (1997). Extremophiles. Sci Am 276, 82‐87. Marx, A., de Graaf, A. A., Wiechert, W., Eggeling, L. and Sahm, H. (1996). Determination of the
fluxes in the central metabolism of Corynebacterium glutamicum by nuclear magnetic resonance spectroscopy combined with metabolite balancing. Biotechnol Bioeng 49, 111‐129.
Marx, A., Hans, S., Möckel, B., Bathe, B., de Graaf, A. A., McCormack, A. C., Stapleton, C., Burke, K.,
O'Donohue, M. and Dunican, L. K. (2003). Metabolic phenotype of phosphoglucose isomerase mutants of Corynebacterium glutamicum. J Biotechnol 104, 185‐197.
McDonald, P., Henderson, A. and Heron, S. J. (1991). The biochemistry of silage. Chalcombe
Publications: Edinburgh. Mecking, S. (2004). Nature or petrochemistry?—Biologically degradable materials. Angew Chem,
Int Ed 43, 1078 –1085. Melzer, G., Eslahpazir, M., Franco‐Lara, E., Nörtemann, B. and Wittmann, C. (2009).
Elementarmoden‐Analyse mit Aspergillus niger zur Optimierung der Sucrase‐Produktion. Chem Ing Tech 81, 1253.
Mimitsuka, T., Sawai, H., Hatsu, M. and Yamada, K. (2007). Metabolic engineering of
Corynebacterium glutamicum for cadaverine fermentation. Biosci Biotechnol Biochem 71, 2130‐2135.
Misono, H. and Soda, K. (1980). Properties of meso‐alpha, epsilon‐diaminopimelate D‐
dehydrogenase from Bacillus sphaericus. J Biol Chem 255, 10599‐10605. Mitchell, H. H. and Smuts, D. B. (1932). The amino acid deficiencies of beef, wheat, corn, oats and
soy beans for growth in the white rat. J Biol Chem 95, 263‐281. Moritz, B., Striegel, K., De Graaf, A. A. and Sahm, H. (2000). Kinetic properties of the glucose‐6‐
phosphate and 6‐phosphogluconate dehydrogenases from Corynebacterium glutamicum and their application for predicting pentose phosphate pathway flux in vivo. Eur J Biochem 267, 3442‐3452.
References
106
Netzer, R., Krause, M., Rittmann, D., Peters‐Wendisch, P. G., Eggeling, L., Wendisch, V. F. and Sahm, H. (2004). Roles of pyruvate kinase and malic enzyme in Corynebacterium glutamicum for growth on carbon sources requiring gluconeogenesis. Arch Microbiol 182, 354‐363.
Neuner, A. and Heinzle, E. (2011). Mixed glucose and lactate uptake by Corynebacterium
glutamicum through metabolic engineering. Biotechnol J 6, 318‐329. Ohnishi, J., Mitsuhashi, S., Hayashi, M., Ando, S., Yokoi, H., Ochiai, K. and Ikeda, M. (2002). A novel
methodology employing Corynebacterium glutamicum genome information to generate a new L‐lysine producing mutant. Appl Microbiol Biotechnol 58, 217‐223.
Okino, S., Noburyu, R., Suda, M., Jojima, T., Inui, M. and Yukawa, H. (2008). An efficient succinic
acid production process in a metabolically engineered Corynebacterium glutamicum strain. Appl Microbiol Biotechnol 81, 459‐464.
Papin, J. A., Stelling, J., Price, N. D., Klamt, S., Schuster, S. and Palsson, B. O. (2004). Comparison of
network‐based pathway analysis methods. Trends Biotechnol 22, 400‐405. Park, J. M., Kim, T. Y. and Lee, S. Y. (2010). Prediction of metabolic fluxes by incorporating
genomic context and flux‐converging pattern analyses. Proc Natl Acad Sci U S A 107, 14931‐6.
Petersen, S., de Graaf, A. A., Eggeling, L., Mollney, M., Wiechert, W. and Sahm, H. (2000). In vivo
quantification of parallel and bidirectional fluxes in the anaplerosis of Corynebacterium glutamicum. J Biol Chem 275, 35932‐35941.
Pfefferle, W., Mockel, B., Bathe, B. and Marx, A. (2003). Biotechnological manufacture of lysine.
Adv Biochem Eng Biotechnol 79, 59‐112. Pfeiffer, T., Sanchez‐Valdenebro, I., Nuno, J. C., Montero, F. and Schuster, S. (1999). METATOOL:
for studying metabolic networks. Bioinformatics 15, 251‐257. Pinstrup‐Andersen, P. and Pandya‐Lorch, R. (1998). Foodsecurity and sustainable use of
naturalresources: a 2020 Vision. Ecological Economics 26, 1‐10. Poolman, M. G., Venkatesh, K. V., Pidcock, M. K. and Fell, D. A. (2004). A method for the
determination of flux in elementary modes, and its application to Lactobacillus rhamnosus. Biotechnol Bioeng 88, 601‐612.
Price, N. D., Papin, J. A. and Palsson, B. O. (2002). Determination of redundancy and systems
properties of the metabolic network of Helicobacter pylori using genome‐scale extreme pathway analysis. Genome Res 12, 760‐769.
Price, N. D., Reed, J. L., Papin, J. A., Famili, I. and Palsson, B. O. (2003). Analysis of metabolic
capabilities using singular value decomposition of extreme pathway matrices. Biophys J 84, 794‐804.
Radhakrishnan, D., Rajvanshi, M. and Venkatesh, K. V. (2010). Phenotypic characterization of
Corynebacterium glutamicum using elementary modes towards synthesis of amino acids. Syst Synth Biol 4, 281‐291.
References
107
Rittmann, D., Lindner, S. N. and Wendisch, V. F. (2008). Engineering of a glycerol utilization pathway for amino acid production by Corynebacterium glutamicum. Appl Environ Microbiol 74, 6216‐6222.
Rittmann, D., Schaffer, S., Wendisch, V. F. and Sahm, H. (2003). Fructose‐1,6‐bisphosphatase from
Corynebacterium glutamicum: expression and deletion of the fbp gene and biochemical characterization of the enzyme. Arch Microbiol 180, 285‐292.
Rowlands, R. T. (1984). Industrial strain improvement: Mutagenesis and random screening
procedures. Enzyme Microb Technol 6, 3‐10. Sahm, H., Eggeling, L. and de Graaf, A. A. (2000). Pathway analysis and metabolic engineering in
Corynebacterium glutamicum. Biol Chem 381, 899‐910. Sahm, H., Eggeling, L., Eikmanns, B. J. and Krämer, R. (1995). Metabolic design in amino acid
producing bacterium Corynebacterium glutamicum. FEMS Microbiol Rev 16, 243‐252. Sauer, U. and Eikmanns, B. J. (2005). The PEP‐pyruvate‐oxaloacetate node as the switch point for
carbon flux distribution in bacteria. FEMS Microbiol Rev 29, 765‐794. Scheer, E., Eggeling, L. and Sahm, H. (1988). Improved D,L‐a‐hydroxybutyrate conversion to L‐
isoleucine with Corynebacterium glutamicum mutants with increased D‐lactate utilization. Appl Microbiol Biotechnol 28, 474‐477.
Schilling, C. H., Edwards, J. S., Letscher, D. and Palsson, B. O. (2000). Combining pathway analysis
with flux balance analysis for the comprehensive study of metabolic systems. Biotechnol Bioeng 71, 286‐306.
Schluesener, D., Fischer, F., Kruip, J., Rogner, M. and Poetsch, A. (2005). Mapping the membrane
proteome of Corynebacterium glutamicum. Proteomics 5, 1317‐1330. Schneider, J., Eberhardt, D. and Wendisch, V. F. (2012). Improving putrescine production by
Corynebacterium glutamicum by fine‐tuning ornithine transcarbamoylase activity using a plasmid addiction system. Appl Microbiol Biotechnol 95, 169‐178.
Schneider, J., Niermann, K. and Wendisch, V. F. (2011). Production of the amino acids l‐glutamate,
l‐lysine, l‐ornithine and l‐arginine from arabinose by recombinant Corynebacterium glutamicum. J Biotechnol 154, 191‐198.
Schneider, J. and Wendisch, V. F. (2010). Putrescine production by engineered Corynebacterium
glutamicum. Appl Microbiol Biotechnol 88, 859‐868. Schneider, K., Schutz, V., John, G. T. and Heinzle, E. (2010). Optical device for parallel online
measurement of dissolved oxygen and pH in shake flask cultures. Bioprocess Biosyst Eng 33, 541‐7.
Schrumpf, B., Schwarzer, A., Kalinowski, J., Puhler, A., Eggeling, L. and Sahm, H. (1991). A
functionally split pathway for lysine synthesis in Corynebacterium glutamicum. J Bacteriol 173, 4510‐4516.
References
108
Schuster, S., Dandekar, T. and Fell, D. A. (1999). Detection of elementary flux modes in biochemical networks: a promising tool for pathway analysis and metabolic engineering. Trends Biotechnol 17, 53‐60.
Schuster, S., von Kamp, A. and Pachkov, M. (2007). Understanding the roadmap of metabolism by
pathway analysis. Methods Mol Biol 358, 199‐226. Schwarzer, A. and Puhler, A. (1991). Manipulation of Corynebacterium glutamicum by gene
disruption and replacement. Nat Biotechnol 9, 84‐87. Seibold, G., Auchter, M., Berens, S., Kalinowski, J. and Eikmanns, B. J. (2006). Utilization of soluble
starch by a recombinant Corynebacterium glutamicum strain: growth and lysine production. J Biotechnol 124, 381‐391.
Seletzky, J. M., Noack, U., Fricke, J., Hahn, S. and Buchs, J. (2006). Metabolic activity of
Corynebacterium glutamicum grown on L: ‐lactic acid under stress. Appl Microbiol Biotechnol 72, 1297‐1307.
Shiio, I., Sugimoto, S. and Kawamura, K. (1990). Effects of carbon source sugars on the yield of
amino acid production and sucrose metabolism in Brevibacterium flavum. Agric Biol Chem. 54, 1513‐1519.
Shiio, I., Sugimoto, S. and Toride, Y. (1984). Studies on mechanisms for lysine production by
pyruvate kinase‐deficient mutants of Brevibacterium flavum. Agric Biol Chem 48, 1551‐1558.
Shinfuku, Y., Sorpitiporn, N., Sono, M., Furusawa, C., Hirasawa, T. and Shimizu, H. (2009).
Development and experimental verification of a genome‐scale metabolic model for Corynebacterium glutamicum. Microb Cell Fact 8, 43.
Sijbesma, F. and Schepens, H. (Eds) (2003). White Biotechnology: Gateway to a more sustainable
future, EuropaBio: Lyon. Smith, K. M., Cho, K. M. and Liao, J. C. (2010). Engineering Corynebacterium glutamicum for
isobutanol production. Appl Microbiol Biotechnol 87, 1045‐1055. Södergard, A., Stolt, M. (2002). Properties of lactic acid based polymers and their correlation with
composition. Prog Polym Sci 27, 1123‐1163. Sonntag, K., Eggeling, L., De Graaf, A. A. and Sahm, H. (1993). Flux partitioning in the split pathway
of lysine synthesis in Corynebacterium glutamicum. Quantification by 13C‐ and 1H‐NMR spectroscopy. Eur J Biochem 213, 1325‐1331.
Spoelstra, S. F., Courtin, M. G. and Van Beers, J. A. C. (1988). Acetic acid bacteria can initiate
aerobic deterioration of maize silage. J Agric Sci Camb 111, 127‐132. Stansen, C., Uy, D., Delaunay, S., Eggeling, L., Goergen, J. L. and Wendisch, V. F. (2005).
Characterization of a Corynebacterium glutamicum lactate utilization operon induced during temperature‐triggered glutamate production. Appl Environ Microbiol 71, 5920‐5928.
References
109
Stelling, J., Klamt, S., Bettenbrock, K., Schuster, S. and Gilles, E. D. (2002). Metabolic network structure determines key aspects of functionality and regulation. Nature 420, 190‐193.
Stephanopoulos, G. (1999). Metabolic fluxes and metabolic engineering. Metab Eng 1, 1‐11. Stephanopoulos, G., Aristidou, A. and Nielsen, J. (1998). Metabolic engineering: principles and
methodologies. Academic Press: San Diego. Stryer, L. (2007). Biochemie. Spektrum Akademischer Verlag: Heidleberg Berlin Oxford. Tateno, T., Fukuda, H. and Kondo, A. (2007). Production of L‐Lysine from starch by
Corynebacterium glutamicum displaying alpha‐amylase on its cell surface. Appl Microbiol Biotechnol 74, 1213‐1220.
Tauch, A., Homann, I., Mormann, S., Ruberg, S., Billault, A., Bathe, B., Brand, S., Brockmann‐
Gretza, O., Ruckert, C., Schischka, N., Wrenger, C., Hoheisel, J., Mockel, B., Huthmacher, K., Pfefferle, W., Puhler, A. and Kalinowski, J. (2002a). Strategy to sequence the genome of Corynebacterium glutamicum ATCC 13032: use of a cosmid and a bacterial artificial chromosome library. J Biotechnol 95, 25‐38.
Tauch, A., Kirchner, O., Löffler, B., Götker, S., Pühler, A. and Kalinowski, J. (2002b). Efficient
electrotransformation of Corynebacterium diphtheriae with a mini‐replicon derived from the Corynebacterium glutamicum plasmid pGA1. Curr Microbiol 45, 362‐367.
Terzer, M. and Stelling, J. (2008). Large‐scale computation of elementary flux modes with bit
pattern trees. Bioinformatics 24, 2229‐2235. Tzvetkov, M., Klopprogge, C., Zelder, O. and Liebl, W. (2003). Genetic dissection of trehalose
biosynthesis in Corynebacterium glutamicum: inactivation of trehalose production leads to impaired growth and an altered cell wall lipid composition. Microbiology 149, 1659‐1673.
Udaka, S. (1960). Screening method for microorganisms accumulating metabolites and its use in
isolation of Micrococcus glutamicus. J Bacteriol 79, 754‐755. Vallino, J. J. and Stephanopoulos, G. (1993). Metabolic flux distributions in Corynebacterium
glutamicum during growth and lysine overproduction. Biotechnol Bioeng 41, 633‐646. van der Rest, M. E. (1999). A heat schock following electroporation induces highly efficient
transformation of Corynebacterium glutamicum with xenogeneic plasmid DNA. Appl Microbiol Biotechnol 52, 541‐545.
van Maris, A. J., Abbott, D. A., Bellissimi, E., van den Brink, J., Kuyper, M., Luttik, M. A., Wisselink,
H. W., Scheffers, W. A., van Dijken, J. P. and Pronk, J. T. (2006). Alcoholic fermentation of carbon sources in biomass hydrolysates by Saccharomyces cerevisiae: current status. Antonie Van Leeuwenhoek 90, 391‐418.
Van Soest, P. J. (1982). Nutritional ecology of the ruminant, 2nd edition edn. Cornell University
Press: Ithaca, New York. Vertes, A. A., Inui, M. and Yukawa, H. (2005). Manipulating corynebacteria, from individual genes
to chromosomes. Appl Environ Microbiol 71, 7633‐7642.
References
110
Wackernagel, M. and Yount, J. D. (2000). Footprints for sustainability: The next steps. Environment, Development and Sustainability 2, 23‐44.
Wehrmann, A., Phillipp, B., Sahm, H. and L.Eggeling (1998). Different modes of diaminopimelate
synthesis and their role in cell wall integrity: a study with Corynebacterium glutamicum. J Bacteriol 180, 3159‐3165.
Weiland, P. (2003). Production and energetic use of biogas from energy crops and wastes in
Germany. Appl Biochem Biotechnol 109, 263‐274. Weinberg, Z. G. and Muck, R., E. (1996). New trends and opportunities in the development and
use of inoculants for silage. FEMS Microbiol Rev 19, 53‐68. Wendisch, V. F., Bott, M., Kalinowski, J., Oldiges, M. and Wiechert, W. (2006). Emerging
Corynebacterium glutamicum systems biology. J Biotechnol 124, 74‐92. Wiback, S. J., Mahadevan, R. and Palsson, B. O. (2004). Using metabolic flux data to further
constrain the metabolic solution space and predict internal flux patterns: the Escherichia coli spectrum. Biotechnol Bioeng 86, 317‐331.
Wiechert, W., Mollney, M., Petersen, S. and de Graaf, A. A. (2001). A universal framework for 13C
metabolic flux analysis. Metab Eng 3, 265‐283. Wiseman, H. G. and Irvin, H. M. (1957). Determination of organic acids in silage. Agric Food Chem
5, 212‐215. Wittmann, C. (2010). Analysis and engineering of metabolic pathway fluxes in Corynebacterium
glutamicum. Adv Biochem Eng Biotechnol 120, 21‐49. Wittmann, C. and Becker, J. (2007). The l ‐Lysine Story: From Metabolic Pathways to Industrial
Production. Microbiology Monographs 5, 39‐70. Wittmann, C. and de Graaf, A. A. (2005). Metabolic flux analysis in Corynebacterium glutamicum,
Eggeling, L. and Bott, M. (Eds), Handbook of Corynebacterium glutamicum, Taylor & Francis Group: Boca Raton, Florida, pp. 277‐304.
Wittmann, C. and Heinzle, E. (1999). Mass spectrometry for metabolic flux analysis. Biotechnol
Bioeng 62, 739‐750. Wittmann, C. and Heinzle, E. (2001a). Application of MALDI‐TOF MS to lysine‐producing
Corynebacterium glutamicum: a novel approach for metabolic flux analysis. Eur J Biochem 268, 2441‐2455.
Wittmann, C. and Heinzle, E. (2001b). Modeling and experimental design for metabolic flux
analysis of lysine‐producing Corynebacteria by mass spectrometry. Metab Eng 3, 173‐191. Wittmann, C. and Heinzle, E. (2002). Genealogy profiling through strain improvement by using
metabolic network analysis: metabolic flux genealogy of several generations of lysine‐producing corynebacteria. Appl Environ Microbiol 68, 5843‐5859.
References
111
Wittmann, C., Kiefer, P. and Zelder, O. (2004). Metabolic fluxes in Corynebacterium glutamicum during lysine production with sucrose as carbon source. Appl Environ Microbiol 70, 7277‐7287.
Wlaschin, A. P., Trinh, C. T., Carlson, R. and Srienc, F. (2006). The fractional contributions of
elementary modes to the metabolism of Escherichia coli and their estimation from reaction entropies. Metab Eng 8, 338‐352.
Woolford, M. K. (1990). The deterimental effects of air on silage. J Appl Bacteriol 68, 101‐116. Yang, T. H., Heinzle, E. and Wittmann, C. (2005). Theoretical aspects of 13C metabolic flux analysis
with sole quantification of carbon dioxide labeling. Comput Biol Chem 29, 121‐133. Yang, T. H., Wittmann, C. and Heinzle, E. (2006). Respirometric 13C flux analysis, Part I: design,
construction and validation of a novel multiple reactor system using on‐line membrane inlet mass spectrometry. Metab Eng 8, 417‐31.
Yu, K. M., Curcic, I., Gabriel, J. and S.C., T. (2008). Recent advances in CO2 capture and utilization.
ChemSusChem 1, 893–899.
Supplementary material
112
Supplementary material
Supplementary material Chapter 3
Supplementary Tables Table S 3.1: List of reactions used for elementary mode analysis (Figure 3.1). Single
arrows – irreversible reactions, double arrows – reversible reactions. Formulation as
used as input for efmtool (http:/www.csb.ethz.ch/tools/efmtool; (Terzer and Stelling,
2008). Reactions are deduced from (Krömer et al., 2004), KEGG (Kyoto
Encyclopedia of Genes and Genomes; http://www.genome.jp/kegg/), (Kjeldsen and
Nielsen, 2009; Melzer et al., 2009; Shinfuku et al., 2009).
R1 : # --> GLCex R2 : # --> LACex R3 : # --> O2 R4 : # --> NH3 R5 : NH3 --> # R6 : # --> CO2 R7 : CO2 --> # R8 : PEP + GLCex --> PYR + G6P R9 : LACex + NAD --> PYR + NADH R10 : G6P <--> F6P R11 : G6P + NADP --> GLC-LAC + NADPH R12 : GLC-LAC --> 6-P-Gluconate R13 : 6-P-Gluconate + NADP --> RIB-5P + CO2 + NADPH R14 : RIB-5P <--> XYL-5P R15 : RIB-5P <--> RIBO-5P R16 : S7P + GA3P <--> RIBO-5P + XYL-5P R17 : S7P + GA3P <--> E-4P + F6P R18 : F6P + GA3P <--> E-4P + XYL-5P R19 : ATP + F6P --> ADP + F-16-BP R20 : F-16-BP --> F6P R21 : F-16-BP <--> GA3P + DAHP R22 : DAHP <--> GA3P R23 : GA3P + NAD <--> 13-PG + NADH R24 : ADP + 13-PG <--> ATP + 3-PG
Supplementary material
113
R25 : 3-PG <--> 2-PG R26 : 2-PG <--> PEP R27 : PEP + ADP --> PYR + ATP R28 : PYR + H-CoA + NAD --> AC-CoA + NADH + CO2 R29 : AC-CoA + OAA --> CIT + H-CoA R30 : CIT <--> Cis-ACO R31 : Cis-ACO <--> ICI R32 : ICI + NADP --> 2-OXO + CO2 + NADPH R33 : 2-OXO + NH3 + NADPH --> GLU + NADP R34 : 2-OXO + NAD + H-CoA --> SUCC-CoA + NADH + CO2 R35 : SUCC-CoA + ADP <--> SUCC + H-CoA + ATP R36 : SUCC + FAD <--> FUM + FADH R37 : FUM <--> MAL R38 : MAL + NAD <--> OAA + NADH R39 : ICI --> GLYOXY + SUCC R40 : GLYOXY + AC-CoA --> MAL + H-CoA R41 : PYR + ATP + CO2 --> OAA + ADP R42 : PEP + CO2 --> OAA R43 : OAA + ATP <--> PEP + ADP + CO2 R44 : OAA + GLU <--> ASP + 2-OXO R45 : ASP + ATP --> ASP-P + ADP R46 : ASP-P + NADPH --> ASP-SA + NADP R47 : ATP --> ADP R48 : MAL + NADP <--> PYR + CO2 + NADPH R49 : 2 NADH + O2 + 4 ADP --> 2 NAD + 4 ATP R50 : 2 FADH + O2 + 2 ADP --> 2 FAD + 2 ATP R51 : 6231 NH3 + 205 G6P + 308 F6P + 879 RIBO-5P + 268 E-4P + 129 GA3P + 1295 3- PG + 652 PEP + 2604 PYR + 3177 AC-CoA + 1680 OAA + 1224 2-OXO + 16429
NADPH + 17002 ATP + 3111 NAD --> 40552 BIOMASS + 16429 NADP + 3177 H-CoA + 1227 CO2 + 17002 ADP + 3111 NADH
R52 : AMP + ATP --> 2 ADP R53 : ASP-SA + PYR --> DHP R54 : DHP + NADPH <--> NADP + PDC R55 : PDC + NH3 + NADPH <--> DMP + NADP R56 : PDC + SUCC-CoA --> SAK + H-CoA R57 : SAK + GLU <--> SDP + 2-OXO R58 : SDP --> DMP + SUCC R59 : DMP --> LYS + CO2 R60 : GLCex + ATP --> G6P + ADP R61 : LYS --> # R62 : # --> AMP R63 : BIOMASS --> #
Supplementary material
114
Table S 3.2: Examples of elementary flux modes with highest L-lysine yields during
growth on lactate (only fluxes with non-zero values are listed). A – mode with highest
yield, B – mode with highest yield not using glucose 6-phosphate dehydrogenase.
A) Mode with highest yield. YLys/S =0.750 (C-mol L-lysine)/(C-mol lactate)
R# : Flux reaction R2 : 4.0 # --> LACex R3 : 1.5 # --> O2 R4 : 3.0 # --> NH3 R7 : 3.0 CO2 --> # R9 : 4.0 LACex + NAD --> PYR + NADH R10 : -3.0 G6P <--> F6P R11 : 3.0 G6P + NADP --> GLC-LAC + NADPH R12 : 3.0 GLC-LAC --> 6-P-Gluconate R13 : 3.0 6-P-Gluconate + NADP --> RIB-5P + CO2 + NADPH R14 : 2.0 RIB-5P <--> XYL-5P R15 : 1.0 RIB-5P <--> RIBO-5P R16 : -1.0 S7P + GA3P <--> RIBO-5P + XYL-5P R17 : 1.0 S7P + GA3P <--> E-4P + F6P R18 : -1.0 F6P + GA3P <--> E-4P + XYL-5P R20 : 1.0 F-16-BP --> F6P R21 : -1.0 F-16-BP <--> GA3P + DAHP R22 : -1.0 DAHP <--> GA3P R23 : -1.0 GA3P + NAD <--> 13-PG + NADH R24 : -1.0 ADP + 13-PG <--> ATP + 3-PG R25 : -1.0 3-PG <--> 2-PG R26 : -1.0 2-PG <--> PEP R33 : 1.5 2-OXO + NH3 + NADPH --> GLU + NADP R41 : 2.5 PYR + ATP + CO2 --> OAA + ADP R43 : 1.0 OAA + ATP <--> PEP + ADP + CO2 R44 : 1.5 OAA + GLU <--> ASP + 2-OXO R45 : 1.5 ASP + ATP --> ASP-P + ADP R46 : 1.5 ASP-P + NADPH --> ASP-SA + NADP R49 : 1.5 2 NADH + O2 + 4 ADP --> 2 NAD + 4 ATP R53 : 1.5 ASP-SA + PYR --> DHP R54 : 1.5 DHP + NADPH <--> NADP + PDC R55 : 1.5 PDC + NH3 + NADPH <--> DMP + NADP R59 : 1.5 DMP --> LYS + CO2 R61 : 1.5 LYS --> #
Supplementary material
115
B) Mode with highest yield not using glucose-6-phosphate dehydrogenase (R11). YLys/S
=0.722 (C-mol L-lysine)/(C-mol lactate)
R# : Flux reaction R2 : 7.2 # --> LACex R3 : 3.4 # --> O2 R4 : 5.2 # --> NH3 R7 : 6.0 CO2 --> # R9 : 7.2 LACex + NAD --> PYR + NADH R28 : 2.0 PYR + H-CoA + NAD --> AC-CoA + NADH + CO2 R29 : 2.0 AC-CoA + OAA --> CIT + H-CoA R30 : 2.0 CIT <--> Cis-ACO R31 : 2.0 Cis-ACO <--> ICI R32 : 2.0 ICI + NADP --> 2-OXO + CO2 + NADPH R33 : 2.6 2-OXO + NH3 + NADPH --> GLU + NADP R34 : 2.0 2-OXO + NAD + H-CoA --> SUCC-CoA + NADH + CO2 R35 : 2.0 SUCC-CoA + ADP <--> SUCC + H-CoA + ATP R36 : 2.0 SUCC + FAD <--> FUM + FADH R37 : 2.0 FUM <--> MAL R38 : -6.4 MAL + NAD <--> OAA + NADH R41 : 11.0 PYR + ATP + CO2 --> OAA + ADP R44 : 2.6 OAA + GLU <--> ASP + 2-OXO R45 : 2.6 ASP + ATP --> ASP-P + ADP R46 : 2.6 ASP-P + NADPH --> ASP-SA + NADP R48 : 8.4 MAL + NADP <--> PYR + CO2 + NADPH R49 : 2.4 2 NADH + O2 + 4 ADP --> 2 NAD + 4 ATP R50 : 1.0 2 FADH + O2 + 2 ADP --> 2 FAD + 2 ATP R53 : 2.6 ASP-SA + PYR --> DHP R54 : 2.6 DHP + NADPH <--> NADP + PDC R55 : 2.6 PDC + NH3 + NADPH <--> DMP + NADP R59 : 2.6 DMP --> LYS + CO2 R61 : 2.6 LYS --> #
Supplementary material
116
Table S 3.3: Examples of elementary flux modes with highest L-lysine yields during
growth on glucose (only fluxes with non-zero values are listed). A – mode with highest
yield including malic enzyme activity, B – mode with highest L-lysine yield including
malic enzyme activity and pts but without glucokinase, C – mode with highest yield
excluding malic enzyme activity and using only pts.
A) mode with highest yield including malic enzyme activity (R48). YLys/S =0.857 (C-mol
L-lysine)/(C-mol glucose)
R# : Flux reaction R1 : 7 # --> GLCex R4 : 12 # --> NH3 R7 : 6 CO2 --> # R10 : 1 G6P <--> F6P R11 : 6 G6P + NADP --> GLC-LAC + NADPH R12 : 6 GLC-LAC --> 6-P-Gluconate R13 : 6 6-P-Gluconate + NADP --> RIB-5P + CO2 + NADPH R14 : 4 RIB-5P <--> XYL-5P R15 : 2 RIB-5P <--> RIBO-5P R16 : -2 S7P + GA3P <--> RIBO-5P + XYL-5P R17 : 2 S7P + GA3P <--> E-4P + F6P R18 : -2 F6P + GA3P <--> E-4P + XYL-5P R19 : 5 ATP + F6P --> ADP + F-16-BP R21 : 5 F-16-BP <--> GA3P + DAHP R22 : 5 DAHP <--> GA3P R23 : 12 GA3P + NAD <--> 13-PG + NADH R24 : 12 ADP + 13-PG <--> ATP + 3-PG R25 : 12 3-PG <--> 2-PG R26 : 12 2-PG <--> PEP R33 : 6 2-OXO + NH3 + NADPH --> GLU + NADP R38 : -12 MAL + NAD <--> OAA + NADH R41 : 6 PYR + ATP + CO2 --> OAA + ADP R43 : -12 OAA + ATP <--> PEP + ADP + CO2 R44 : 6 OAA + GLU <--> ASP + 2-OXO R45 : 6 ASP + ATP --> ASP-P + ADP R46 : 6 ASP-P + NADPH --> ASP-SA + NADP R48 : 12 MAL + NADP <--> PYR + CO2 + NADPH R53 : 6 ASP-SA + PYR --> DHP R54 : 6 DHP + NADPH <--> NADP + PDC R55 : 6 PDC + NH3 + NADPH <--> DMP + NADP R59 : 6 DMP --> LYS + CO2 R60 : 7 GLCex + ATP --> G6P + ADP
Supplementary material
117
R62 : 6 LYS --> #
B) mode with highest L-lysine yield including malic enzyme (R48) and pts (R8) activity
but no glucokinase (R60) activity. YLys/S =0.833 (C-mol L-lysine)/(C-mol glucose)
R# : Flux reaction R1 : 6.0 # --> GLCex R3 : 1.0 # --> O2 R4 : 10.0 # --> NH3 R7 : 6.0 CO2 --> # R8 : 6.0 PEP + GLCex --> PYR + G6P R11 : 6.0 G6P + NADP --> GLC-LAC + NADPH R12 : 6.0 GLC-LAC --> 6-P-Gluconate R13 : 6.0 6-P-Gluconate + NADP --> RIB-5P + CO2 + NADPH R14 : 4.0 RIB-5P <--> XYL-5P R15 : 2.0 RIB-5P <--> RIBO-5P R16 : -2.0 S7P + GA3P <--> RIBO-5P + XYL-5P R17 : 2.0 S7P + GA3P <--> E-4P + F6P R18 : -2.0 F6P + GA3P <--> E-4P + XYL-5P R19 : 4.0 ATP + F6P --> ADP + F-16-BP R21 : 4.0 F-16-BP <--> GA3P + DAHP R22 : 4.0 DAHP <--> GA3P R23 : 10.0 GA3P + NAD <--> 13-PG + NADH R24 : 10.0 ADP + 13-PG <--> ATP + 3-PG R25 : 10.0 3-PG <--> 2-PG R26 : 10.0 2-PG <--> PEP R33 : 5.0 2-OXO + NH3 + NADPH --> GLU + NADP R38 : -8.0 MAL + NAD <--> OAA + NADH R41 : 9.0 PYR + ATP + CO2 --> OAA + ADP R43 : -4.0 OAA + ATP <--> PEP + ADP + CO2 R44 : 5.0 OAA + GLU <--> ASP + 2-OXO R45 : 5.0 ASP + ATP --> ASP-P + ADP R46 : 5.0 ASP-P + NADPH --> ASP-SA + NADP R48 : 8.0 MAL + NADP <--> PYR + CO2 + NADPH R49 : 1.0 2 NADH + O2 + 4 ADP --> 2 NAD + 4 ATP R53 : 5.0 ASP-SA + PYR --> DHP R54 : 5.0 DHP + NADPH <--> NADP + PDC R55 : 5.0 PDC + NH3 + NADPH <--> DMP + NADP R59 : 5.0 DMP --> LYS + CO2 R62 : 5.0 LYS --> #
Supplementary material
118
Table S 3.4: Examples of elementary flux modes with highest L-lysine yields during
growth with simultaneous consumption of glucose and lactate (only fluxes with non-
zero values are listed). A – mode with highest yield mode including malic enzyme
activity, B – mode highest yield including malic enzyme activity and pts but without
glucokinase, C – mode with highest yield excluding malic enzyme activity and using
only pts.
A) mode with highest yield mode including malic enzyme activity (R48). YLys/S =0.848 (C-
mol L-lysine)/(C-mol glucose)
R# : Flux reaction R1 : 15.0 # --> GLCex R2 : 3.0 # --> LACex R3 : 1.0 # --> O2 R4 : 28.0 # --> NH3 R7 : 15.0 CO2 --> # R9 : 3.0 LACex + NAD --> PYR + NADH R11 : 15.0 G6P + NADP --> GLC-LAC + NADPH R12 : 15.0 GLC-LAC --> 6-P-Gluconate R13 : 15.0 6-P-Gluconate + NADP --> RIB-5P + CO2 + NADPH R14 : 10.0 RIB-5P <--> XYL-5P R16 : -5.0 S7P + GA3P <--> RIBO-5P + XYL-5P R17 : 5.0 S7P + GA3P <--> E-4P + F6P R18 : -5.0 F6P + GA3P <--> E-4P + XYL-5P R19 : 10.0 ATP + F6P --> ADP + F-16-BP R21 : 10.0 F-16-BP <--> GA3P + DAHP R22 : 10.0 DAHP <--> GA3P R23 : 25.0 GA3P + NAD <--> 13-PG + NADH R24 : 25.0 ADP + 13-PG <--> ATP + 3-PG R25 : 25.0 3-PG <--> 2-PG R26 : 25.0 2-PG <--> PEP R33 : 14.0 2-OXO + NH3 + NADPH --> GLU + NADP R38 : -26.0 MAL + NAD <--> OAA + NADH R41 : 15.0 PYR + ATP + CO2 --> OAA + ADP R43 : -25.0 OAA + ATP <--> PEP + ADP + CO2 R44 : 14.0 OAA + GLU <--> ASP + 2-OXO R45 : 14.0 ASP + ATP --> ASP-P + ADP R46 : 14.0 ASP-P + NADPH --> ASP-SA + NADP R48 : 26.0 MAL + NADP <--> PYR + CO2 + NADPH R49 : 1.0 2 NADH + O2 + 4 ADP --> 2 NAD + 4 ATP R53 : 14.0 ASP-SA + PYR --> DHP R54 : 14.0 DHP + NADPH <--> NADP + PDC
Supplementary material
119
R55 : 14.0 PDC + NH3 + NADPH <--> DMP + NADP R59 : 14.0 DMP --> LYS + CO2 R60 : 15.0 GLCex + ATP --> G6P + ADP R62 : 14.0 LYS --> # B) mode highest yield including malic enzyme (R48) and pts (R8) activity but no
glucokinase (R60) activity. YLys/S =0.792 (C-mol L-lysine)/(C-mol glucose+C-mol lactate)
R# : Flux reaction R1 : 1.0 # --> GLCex R2 : 2.8 # --> LACex R3 : 1.1 # --> O2 R4 : 3.8 # --> NH3 R7 : 3.0 CO2 --> # R8 : 1.0 PEP + GLCex --> PYR + G6P R9 : 2.8 LACex + NAD --> PYR + NADH R10 : -2.0 G6P <--> F6P R11 : 3.0 G6P + NADP --> GLC-LAC + NADPH R12 : 3.0 GLC-LAC --> 6-P-Gluconate R13 : 3.0 6-P-Gluconate + NADP --> RIB-5P + CO2 + NADPH R14 : 2.0 RIB-5P <--> XYL-5P R15 : 1.0 RIB-5P <--> RIBO-5P R16 : -1.0 S7P + GA3P <--> RIBO-5P + XYL-5P R17 : 1.0 S7P + GA3P <--> E-4P + F6P R18 : -1.0 F6P + GA3P <--> E-4P + XYL-5P R23 : 1.0 GA3P + NAD <--> 13-PG + NADH R24 : 1.0 ADP + 13-PG <--> ATP + 3-PG R25 : 1.0 3-PG <--> 2-PG R26 : 1.0 2-PG <--> PEP R33 : 1.9 2-OXO + NH3 + NADPH --> GLU + NADP R38 : -1.6 MAL + NAD <--> OAA + NADH R41 : 3.5 PYR + ATP + CO2 --> OAA + ADP R44 : 1.9 OAA + GLU <--> ASP + 2-OXO R45 : 1.9 ASP + ATP --> ASP-P + ADP R46 : 1.9 ASP-P + NADPH --> ASP-SA + NADP R48 : 1.6 MAL + NADP <--> PYR + CO2 + NADPH R49 : 1.1 2 NADH + O2 + 4 ADP --> 2 NAD + 4 ATP R53 : 1.9 ASP-SA + PYR --> DHP R54 : 1.9 DHP + NADPH <--> NADP + PDC R55 : 1.9 PDC + NH3 + NADPH <--> DMP + NADP R59 : 1.9 DMP --> LYS + CO2 R62 : 1.9 LYS --> #
Supplementary material
120
C) Mode with highest yield excluding malic enzyme activity (R48) and without
glucokinase (R60) activity. YLys/S =0.75 (C-mol L-lysine)/(C-mol glucose+C-mol lactate)
R# : Flux reaction R1 : 1.0 # --> GLCex R2 : 2.0 # --> LACex R3 : 1.5 # --> O2 R4 : 3.0 # --> NH3 R7 : 3.0 CO2 --> # R8 : 1.0 PEP + GLCex --> PYR + G6P R9 : 2.0 LACex + NAD --> PYR + NADH R10 : -2.0 G6P <--> F6P R11 : 3.0 G6P + NADP --> GLC-LAC + NADPH R12 : 3.0 GLC-LAC --> 6-P-Gluconate R13 : 3.0 6-P-Gluconate + NADP --> RIB-5P + CO2 + NADPH R14 : 2.0 RIB-5P <--> XYL-5P R15 : 1.0 RIB-5P <--> RIBO-5P R16 : -1.0 S7P + GA3P <--> RIBO-5P + XYL-5P R17 : 1.0 S7P + GA3P <--> E-4P + F6P R18 : -1.0 F6P + GA3P <--> E-4P + XYL-5P R23 : 1.0 GA3P + NAD <--> 13-PG + NADH R24 : 1.0 ADP + 13-PG <--> ATP + 3-PG R25 : 1.0 3-PG <--> 2-PG R26 : 1.0 2-PG <--> PEP R27 : 4.0 PEP + ADP --> PYR + ATP R33 : 1.5 2-OXO + NH3 + NADPH --> GLU + NADP R41 : 5.5 PYR + ATP + CO2 --> OAA + ADP R42 : 0.0 PEP + CO2 --> OAA R43 : 4.0 OAA + ATP <--> PEP + ADP + CO2 R44 : 1.5 OAA + GLU <--> ASP + 2-OXO R45 : 1.5 ASP + ATP --> ASP-P + ADP R46 : 1.5 ASP-P + NADPH --> ASP-SA + NADP R49 : 1.5 2 NADH + O2 + 4 ADP --> 2 NAD + 4 ATP R53 : 1.5 ASP-SA + PYR --> DHP R54 : 1.5 DHP + NADPH <--> NADP + PDC R55 : 1.5 PDC + NH3 + NADPH <--> DMP + NADP R59 : 1.5 DMP --> LYS + CO2 R61 : 1.5 LYS --> #
Supplementary material
121
Supplementary Figures
DHP
R53PYR
NADP+
NADPHR54
PDCSucc-CoA
H-CoA
SAK
R56
R57GLU
2-OXO
SDP
SUCC
DMP
R58
R55
+ NADPH
NADP
LYS
CO2 ex
NH3 ex
R59
LYSexR60
ADP
ATP
PEP
ADP ATP
PYR
13-PG
NAD+
NADH
3-PG
2-PG
G6P GLC-LAC
Glcex
F6P
NADP+ NADPH NADP+ NADPH
CO2 exRIB-5P
XYL-5P RIBO-5P
S7P GA3P
E-4P F6PGA3P
F-1,6-BP
ATP
ADP
DHAP
+
NADHNAD+ + H-CoA
OAA
MAL
SUCC
2-OXO
ICI
SUCC-CoA
Ac-CoA
NH3 ex
Glu
CIT
Cis-ACO
FUM
H-CoAATP ADP
H-CoA
NADP+
NADPH
NAD+
NADH
CoA
FADH
FAD+
NAD+
NADH
NADP+NADPH
CO2 ex
CO2 ex+
CO2 ex
ASP
2-OXO
ASP-P
ADP
ATP
ASP-SA
NADPH
NADP+
GLU
CO2 ex
CO2 ex
BIOMASSex
BIOMASSex
BIOMASSex
PEP
PYR
GLYOXY
H-CoA
BIOMASSex
BIOMASSex
R8
R10R11 R12 R13
R14 R15
R51
R51
R16
R7
R18
R19
R51
R51
R51
ATP
CO2 ex
R43 R42
R21
R20
R22
R23
R24
R25R26
R27 R28
R29R30
R31
R32
R33
R34
R35
R36
R37
R38
R39R40
PYR
NADP+
NADPH
CO2 ex
R48
R41ADP
R44
R46
R45
+ ATP
ATP
6-P-Gluconate
Lacex
NADH
R9
R51 : 6231 NH3 ex + 205 G6P +308 F6P + 879 RIBO-5P + 268 E4P + 129 GA3P + 1295 3-PG + 652 PEP + 2604 PYR + 3177 AC-CoA + 1680 OAA + 1224 2-OXO + 16429 NADPH + 17002 ATP + 3111 NAD = 40552 BIOMASSex + 16429 NADP + 3177 H-CoA + 17002 ADP + 3111 NADH + 1227 CO2 ex
O2 exR50 : 2 FADH + + 2 ADP = 2 FAD + 2 ATP
R47 : ATP = ADP
R52 : AMP + ATP = 2 ATP
R49 : 2 NADH + + 4 ADP = 2 NAD + 4 ATPO2 ex
ATP
ADPR60
Figure S 3.1: Metabolic network for the growth and production of biomass using
glucose and lactate as carbon sources (List of reactions is provided in supplementary
Table S 3.1).
Supplementary material
122
Figure S 3.2: Relative carbon flux through malic enzyme (R48) and glucose-6-
phosphate dehydrogenase (R11) and function of relative carbon flux to L-lysine
(R61). Fluxes are related to total carbon uptake rate. Lac –growth only on lactate,
Glc – growth only on glucose, LacGlc – modes with simultaneous uptake of lactate
and glucose, all – all possible substrate variations cumulated.
R48
Cm
ol/C
mol
R48
Cm
ol/C
mol
R48
Cm
ol/C
mol
R48
Cm
ol/C
mol
R11
Cm
ol/C
mol
R11
Cm
ol/C
mol
R11
Cm
ol/C
mol
R11
Cm
ol/C
mol
Supplementary material
123
0 5 10 15 20 25 30 35 40 45 500
0.5
1Y
Lys/
S [
C-m
ol C
-mol
-1]
OF
0 5 10 15 20 25 30 35 40 45 500
0.5
1
YB
M/S
[C
-mol
C-m
ol-1
]
OF
Figure S 3.3: Carbon yields of L-lysine, YLys/S, and biomass, YBM/S, given in C-moles
per C-mol as a function of the objective function, OF, specified in Equation 1.
Supplementary material
124
0 500 10000
50
Lo
ad
-R4
1
Mode0 500 1000
0
20
40
Lo
ad
-R4
3
Mode0 500 1000
0
50
Lo
ad
-R4
8
Mode
0 500 10000
50
Lo
ad
-R4
2
Mode0 500 1000
-20
0
20
40
Lo
ad
-R1
0
Mode0 500 1000
0
50
Lo
ad
-R1
1
Mode
0 500 10000
50
Lo
ad
-R2
6
Mode0 500 1000
0
50
Lo
ad
-R3
2
Mode0 500 1000
0
50
Lo
ad
-R3
9Mode
Figure S 3.4: Elementary modes ranked with increasing objective function, OF, and
selected cumulative enzyme loadings as defined in Equation 2 for modes with growth
on lactate, glucose or mixtures thereof. Reaction numbers are defined in
supplementary Figure S 3.1 and supplementary Table S 3.1.
Supplementary material
125
0 500 10000
50
Lo
ad
-R4
1
Mode0 500 1000
0
20
40
Lo
ad
-R4
3
Mode0 500 1000
0
50
Lo
ad
-R4
8
Mode
0 500 10000
50
Lo
ad
-R4
2
Mode0 500 1000
0
20
40
Lo
ad
-R1
0
Mode0 500 1000
0
50
Lo
ad
-R1
1
Mode
0 500 10000
50
Lo
ad
-R2
6
Mode0 500 1000
0
50
Lo
ad
-R3
2
Mode0 500 1000
0
50
Lo
ad
-R3
9
Mode
Figure S 3.5: Elementary modes ranked with increasing objective function, OF, and
selected cumulative enzyme loadings as defined in Equation 2 for modes with only
uptake of glucose. Reaction numbers are defined in Figure S 3.1 and supplementary
Table S 3.1.
Supplementary material
126
0 500 10000
50
Lo
ad
-R4
1
Mode0 500 1000
0
20
40
Lo
ad
-R4
3
Mode0 500 1000
0
50
Lo
ad
-R4
8
Mode
0 500 10000
50
Lo
ad
-R4
2
Mode0 500 1000
-20
0
20
40
Lo
ad
-R1
0
Mode0 500 1000
0
50
Lo
ad
-R1
1
Mode
0 500 10000
50
Lo
ad
-R2
6
Mode0 500 1000
0
50
Lo
ad
-R3
2
Mode0 500 1000
0
50
Lo
ad
-R3
9Mode
Figure S 3.6: Elementary modes ranked with increasing objective function, OF, and
selected cumulative enzyme loadings as defined in Equation 2 for modes with
simultaneous uptake of glucose and lactate. Reaction numbers are defined in Figure S
3.1 and supplementary Table S 3.1.
Supplementary material
127
Figure S 3.7: Cultivation profile of C. glutamicum ATCC 13032 lysCfbr on D,L-lactate
0 5 10 15 20 25 30 35 40 45 50 55 600
1
2
3
4
5
6 O
D66
0
Time [h]
Supplementary material
128
0
50
100
150
200
250
0 50 100 150 200 250
Glucose [C‐mM]
Lactate [C‐mM]
0
50
100
150
200
250
0 50 100 150 200
Lactate [C‐m
M]
CDW [C‐mM]
0
50
100
150
200
250
0 50 100 150 200
Glucose [C‐mM]
CDW [C‐mM]
Figure S 3.8: Yield plots of growth and L-lysine production of C. glutamicum ATCC
13032 lysCfbrdldPsodpycPsodmalEPsod on a mixture of D, L-lactate and glucose as depicted
in Figure 3.5. A – glucose and lactate concentrations; B – cell dry weight, CDW and
lactate; C – cell dry weight and glucose.
Supplementary material
129
Supplementary material Chapter 4
Supplementary Figures
Figure S 4.1: Batch cultivation of L-lysine producing C. glutamicum SL on 1:4 diluted
grass silage juice in the bioreactor using impeller 2 at a stirring speed of 600 rpm and
a constant air flow rate of 124 ml min-1 (0.124 vvm) and additional oxygenation with
Supplementary material
130
pure oxygen as in (C). (A) Concentrations of substrates fructose (Fru), glucose (Glu)
and lactate (Lac), of product L-lysine (Lys) and of biomass provided as cell dry
weight (CDW). (B) Oxygen uptake rate (OUR) and carbon dioxide production rate
(CPR). (C) Optical density (OD) and dissolved oxygen concentration (DO – full line)
given in % air saturation. Additional pure oxygen flow rate: 1 – 0 ml min-1, 2 –16 ml
min-1, 3 –36 ml min-1, 4 – 76 ml min-1 ml min-1, 5 – 116 ml min-1. Dashed lines indicate
identified growth phases I, II and III.
Figure S 4.2: Profile of shake flask cultivation of L-lysine producing C. glutamicum
SL on grass silage juice. (A) Concentrations of substrates fructose (Fru), glucose (Glu)
and lactate (Lac), of product L-lysine (Lys) and of biomass provided as cell dry
weight (CDW). (B) Optical density (OD) and dissolved oxygen concentration (DO –
0 5 10 15 20
0
1
2
3
4
0
20
40
60
80
100
ln O
D66
0
Time [h]
DO
[%]
0
1
2
3
4
5
6
7
8
0
10
20
30
40A
B
Supplementary material
131
full line) given in % air saturation. Shaking rate: 230 rpm. Dashed lines indicate
identified growth phases I and II.
Figure S 4.3: Profile of shake flask cultivation of L-lysine producing C. glutamicum
SL on corn silage juice. (A) Concentrations of substrates maltose (Mal), glucose (Glu)
and lactate (Lac), of product L-lysine (Lys) and of biomass provided as cell dry
weight (CDW). (B) Optical density (OD) and dissolved oxygen concentration (DO –
full line) given in % air saturation. Shaking rate: 230 rpm. Dashed lines indicate
identified growth phases I and II.
0 5 10 15 20 25 30-1
0
1
2
3
4
0
20
40
60
80
100
lnO
D 660
Time [h]
I II
0
1
2
3
4
5
6
7
0
5
10
15
20
25
Lys
[g/L
]
CD
W [g
/ L]
Lac
[g /
L]M
al [g
/ L]
Glu
[g /
L]
A
B
Curriculum vitae
132
Curriculum vitae
Andreas Neuner
Diplom-Biologe, geboren am 29.06.1975 in Timisoara, Romania
Aktuelle Tätigkeit
Wissenschaftlicher Mitarbeiter seit 04/2012
Einrichten des Lehrstuhls Systembiologie:
Kooperationen mit anderen Arbeitsgruppen
der UdS sowie mit anderen Universitäten
(Université de la Grande Region, UniGR)
Akademische Ausbildung
Promotion 11/2008-11/2012
Institut für technische Biochemie,
Universität des Saarlandes
Titel: “Metabolic engineering of
Corynebacterium glutamicum for L-lysine
production on silage”
Diplomarbeit 10/2007-08/2008 Institut für Mikrobiologie, Universität des
Saarlandes
Titel: „Isolierung und Charakterisierung der
L-Glucitol-Dehydrogenase aus dem
Bakterium Stenotrophomonas maltophilia“
Studium
2002-2008
Universität des Saarlandes,
Universitätsklinikum Homburg/Saar,
Zentrum für Human und Molekularbiologie
Human und Molekularbiologie
1998-2002 Technische Universität Karlsruhe
Studienfach Biologie, Vordiplom
Curriculum vitae
133
Beruflicher Werdegang
1999-2008
ATW Antriebstechnik : Verkauf
1996-1998 Walter Telemarketing & Vertrieb GmbH :
Verkaufsservice, Kundenberatung und
Betreuung
Schulische Ausbildung
1990-1996
Otto Hahn Gymnasium, Gernsbach
Abitur: Biologie, Englisch
1982-1989 Gesamtschule: Scoala generala 16,
Timisoara, Romania
Curriculum vitae
134
Publikationen
Neuner, A., Wagner, I., Sieker, T., Ulber, R., Schneider, K., Peifer, S., Heinzle, E. (2012).
Production of L-lysine on different silage juices using genetically engineered
Corynebacterium glutamicum. J Biotechnol , In press.
Pauling, J., Röttger, R., Neuner, A., Salgado, H., Collado-Vides, J., Kalaghatgi, P.,
Azevedo, V., Tauch, A., Pühler, A., Baumbach, J. (2012). On the trail of EHEC/EAEC-
unraveling the gene regulatory networks of human pathogenic Escherichia coli bacteria.
Integr Biol (Camb) 4, 728-733.
Neuner, A. and Heinzle, E. (2011). Mixed glucose and lactate uptake by Corynebacterium
glutamicum through metabolic engineering. Biotechnol J 6, 318-329
Sieker, T., Dimitrova, D., Neuner, A., Tippkötter, N., Muffler, K., Bart, H.-J., Heinzle, E.,
Ulber, R. (2011). Ethanol production from grass silage by Simultaneous Pretreatment,
Saccharification and Fermentation: First Steps in the Process Development. Eng. Life Sci.
11, 436-442.
Sieker, T., Dimitrova, D., Neuner, A., Tippkötter, N., Bart, H.-J., Heinzle, E., Ulber, R.
(2010). Grassilage als Rohstoff für die chemische Industrie. Chem. Ing. Tech. 82, 1153-
1159.
Sieker, T., Dimitrova, D., Neuner, A., Tippkötter, N., Bart, H.-J., Heinzle, E., Ulber, R.
(2009). Silage - Fermentationsrohstoff für die chemische Industrie? Labor & More 2/2009.
Curriculum vitae
135
Konferenzbeiträge
Neuner, A. and Heinzle, E.: Engineering Mixed Substrate Uptake by C. glutamicum.
Gothenburg Life Science Conference XI: Industrial Systems Biology. August 19-20, 2010,
Gothenburg, Sweden.
Neuner, A. and Heinzle, E.: Production of L-lysine from Silage Using Engineered Strains
of Corynebacterium glutamicum. 7th International Conference on Renewable Resources
and Biorefineries (RRB 7) on June 8-10, Bruges, Belgium
Sieker, T., Dimitrova, D., Neuner, A., Tippkötter, N., Bart, H.-J., Heinzle, E., Ulber, R.:
Produktion von Grundchemikalien aus Silage. DECHEMA Vortrags- und
Diskussionstagung Biokatalyse: Neue Verfahren, neue Produkte 2009.
Sieker, T., Dimitrova, D., Neuner, A., Tippkötter, N., Bart, H.-J., Heinzle, E., Ulber, R.:
Nutzung von Silage zur fermentativen Produktion von Grund- und Feinchemikalien. 27.
Jahrestagung der Biotechnologen.
Sieker, T., Dimitrova, D., Neuner, A., Tippkötter, N., Bart, H.-J., Heinzle, E., Ulber, R.:
An integrated process for the simultanious pretreatment, saccharification and fermentation
of grass silage lignocelluloses. Jahrestreffen der deutschen Katalytiker 2010.
Sieker, T., Dimitrova, D., Neuner, A., Tippkötter, N., Bart, H.-J., Heinzle, E., Ulber, R.:
Prozessintegration in der grünen Bioraffinerie: Das simultanious pretreatment,
saccharification and fermentation Verfahren. DECHEMA Vortrags- und
Diskussionstagung Bioprozessorientiertes Anlagendesign.
Sieker, T., Dimitrova, D., Neuner, A., Tippkötter, N., Bart, H.-J., Heinzle, E., Ulber, R.:
Grüne Bioraffinerie: Ganzheitliche Nutzung von Grassilage für die Herstellung von Grund-
und Feinchemikalien. 28. Jahrestagung der Biotechnologen / ProcessNet 2010.
Sieker, T., Dimitrova, D., Neuner, A., Tippkötter, N., Bart, H.-J., Heinzle, E., Ulber, R.: A
Green Biorefinery for the Production of Ethanol, Organic Acids and Polyphenols from
Grass Silage. 1st European Congress of Applied Biotechnology.
Curriculum vitae
136
Poster
Neuner, A. and Heinzle, E.: Increased D-lactate utilization in Corynebacterium
glutamicum. Sixth International Conference on Renewable Resources & Biorefineries, 7 –
9 June 2010, Düsseldorf
Neuner, A. and Heinzle, E.: Metabolic engineering of Corynebacterium glutamicum for L-
lysine production from silage. 27. DECHEMA Jahrestagung der Biotechnologen, 8. bis 10.
September 2009, Mannheim
Sieker, T., Dimitrova, D., Neuner, A., Tippkötter, N., Bart, H.-J., Heinzle, E., Ulber, R.:
Utilization of silage as fermentation substrate for the chemical industry. 5th Conference on
Renewable Resources and Biorefineries.
Sieker, T., Duwe, A., Dimitrova, D., Neuner, A., Muffler, K., Tippkötter, N., Bart, H.-J.,
Heinzle, E., Ulber, R.: A Silage-Based Circular Flow Green Biorefinery. Industrial Use of
Renewable resources: Chemistry, Biotechnology, Process Engineering.
Sieker, T., Duwe, A., Dimitrova, D., Neuner, A., Muffler, K., Tippkötter, N., Bart, H.-J.,
Heinzle, E., Ulber, R.: Production of Organic and Phenolic Acids in a Circular Flow Green
Biorefinery. 8th Conference on Renewable Resources and Biorefineries.